*Article* **Forestry Policy, Conservation Activities, and Ecosystem Services in the Remote Misuku Hills of Malawi**

#### **Christopher Coutts, Tisha Holmes and April Jackson**


Received: 29 October 2019; Accepted: 18 November 2019; Published: 21 November 2019 -

**Abstract:** Research Highlights: Most of Malawi's land area has been deforested; however, expansive indigenous forests remain in the remote Misuku Hills in Malawi's northern region. Despite its conservation potential, this region of Malawi has been overlooked in forestry conservation research. Background and Objectives: The Misuku Hills is one the most floristically diverse regions in Malawi, but this region is facing similar pressures and forestry policy enforcement challenges that drive deforestation of other regions. This study therefore addresses the questions: What are the forestry policy challenges and opportunities for forest conservation in Malawi? What conservation activities are taking place in the Misuku Hills in support of these policies? What ecosystem services are residents using that are in need of protection? Materials and Methods: A comprehensive inventory and review of the national forest policies and current programs in the Misuku Hills region was compiled through document reviews and communications with governmental and non-governmental stakeholders. A Photovoice exercise was conducted with residents of Chikutu village to create an inventory of resident-identified ecosystem services. Results: While there is an impressive array of policies in place to protect the forests of Malawi, there is little institutionalization or enforcement of these policies. There have been funded conservation programs in the Misuku Hills, but these have been limited to the areas surrounding the three small public forest reserves. The Photovoice exercise revealed that residents rely on an abundance of forest ecosystem services to support their livelihoods, including food, medicine, and timber products. Conclusions: The challenges to conserving forests and their ecosystem services are being met at a local level in a variety of creative ways in the Misuku Hills (e.g., tree planting, beekeeping) that could be used as community-based models for other areas in Africa and elsewhere, where people depend directly on these services to meet daily needs.

**Keywords:** forests; environment; Malawi; ecosystem services; Photovoice; conservation; policy; community-based forest management; participatory forest management

#### **1. Introduction**

#### *1.1. Rates of Deforestation in Malawi*

Deforestation rates in the tropics are increasing after a promising lag supported by conservation efforts and good governance measures during the 1990s and early 2000s [1]. Increased deforestation is particularly concerning in Africa, where land is rapidly converted to support growing populations and expansion of small-scale and subsistence agriculture [2–4]. In the southeast African nation of Malawi, concerns are growing around the preservation of forest resources, the livelihoods dependent on these resources, and the dynamics which threaten their viability. The most methodologically forthcoming

and rigorous estimate available of the forested land area in Malawi reveals that 26.8% of this small African country is forested (Figure 1), and Malawi experienced a 1.6% aggregated loss of tree cover between 1990 and 2010 (1.4% 1990–2000; 0.2% 2000–2010) [5]. The 1.6% total forest loss over this 20-year period is almost completely explained by the equivalent expansion of agricultural land [5]. Most of the agricultural expansion in Malawi has happened on customary land (defined below), outside the jurisdiction of government-controlled parks and reserves [6].

**Figure 1.** Vegetation map of Malawi. Source: Adapted from FAO (2013).

There are many conflicting estimates of Malawi's forest cover, land tenure regimes in forested areas, and overall deforestation rates. These discrepancies and their potential implications are presented in Appendix A. The deforestation rates above are much less grave than other widely cited estimates. Nonetheless, deforestation continues and the persistent drivers of deforestation provide reason to believe that its consequences will continue to threaten the livelihoods of Malawians who are heavily dependent on forest ecosystem services [7].

The available estimates for the land tenure of forested areas indicate a nearly equal distribution in national parks and game reserves, forest reserves, and on customary land [6,8]. National parks, game reserves, and forest reserves are public, government-controlled lands. Customary land is under Tradition Authority, an indigenous geo-political and socio-economic jurisdiction with customary sovereignty under authority of a Chief [9]. Essentially, all land that is not public or privately owned is customary, and customary land accounts for approximately 85% of the total land area in Malawi [10].

#### *1.2. Drivers of Deforestation*

The main drivers of deforestation in Malawi are a rapidly growing and extremely poor population converting forested land to support small-scale subsistence agriculture for food provisioning and income [11] and using wood as a primary energy source.

First, population. The total population of Malawi grew by 35% between 2008 (13,029,498) and 2018 (17,563,749). While this may seem startling, the gross increase in population equates to a current population density of 186 persons/km2 in this comparatively small African country [12]. There are inconsistencies in various country rankings of population density, but 186 persons/km2 currently places Malawi at approximately the 75th most densely populated country in the world, still far less dense than other small African countries (e.g., Rwanda, Burundi, The Gambia), but also less dense than some comparatively larger African countries (e.g., Nigeria, Uganda). A major difference between Malawi is that 84% of the population resides in rural areas, making Malawi one of the least urbanized countries in Africa [12,13]. The population is dispersed, placing greater pressures on the entirety of the landscape. Population density in Malawi has been shown to be a significant socio-economic explanatory variable of deforestation in Malawi [14], and population growth is a perceived underlying driver of land use and land cover change in Malawi [15].

Second, agriculture. Malawi's economy and the livelihood of Malawians are heavily dependent on agriculture, which "accounts for about 36% of the Gross Domestic Product (GDP), 87% of the total employment, and supplies more than 65% of the manufacturing sector's raw materials" [13] (p. 4). A 2009 report estimates an expansion of agriculture by 31% between 1975 and 1990, with the majority of expansion coming through the clearing of indigenous forest and woodlands. The report attributes these changes to the need to feed a rapidly growing population and desires to promote economic growth through expansion of agriculture production [7]. Indigenous forests on customary land have been the main source of additional agricultural land [16]. The overwhelmingly rural population derives its livelihood from small land holdings of 0.5–2 ha per household, but it is not agriculture on these small landholdings per se that is the threat to forests, but rather "poor husbandry techniques in the absence of alternative economic opportunities" [7] (Preface).

Third, poverty. Poverty is a critical and compounding factor which exacerbates the pressures of a growing population primarily dependent on subsistence agriculture. Malawi is the second poorest county in the world, as measured by annual per capita income [17]. A lack of income-generating opportunities leads many Malawians to extract natural resources directly from the immediate environment, both for daily household use and to sell for profit. Despite positive attitudes towards tree planting [18,19] and awareness of the negative consequences of a lack of social support for deforestation—making the intention to cut down forest trees generally low—extreme poverty and a lack of alternative income opportunities [20] continue to fuel deforestation.

Fourth, energy. The extraction of resources from forests is the main source of many essential products for the rural poor, most notably fuelwood. In fact, "over 90% of the energy demands of the country for domestic and industrial use are met from wood energy" [7] (p. 2). Trade in firewood and charcoal is the primary source of income for many rural poor and the only form of cooking fuel for 99% of the population [21].

Other extractive industries supported by forest resources also include poles and timber for home construction and hand tools, as well as non-timber forest products such as thatch, mushrooms, caterpillars, bushmeat, beeswax, medicinal plants, and materials for handicrafts [13,16]. Malawi also has a large pool (in terms of both diversity and quantity) of indigenous fruit tree resources [22–24]. These forest food "resources are an insurance against hunger and malnutrition. They provide much needed dietary diversity which avails both macro and micro nutrients necessary for good health" [13] (p. 30).

Although these goods are "free" to Malawians with access to forests, monetary value can be assigned to these products based on what a person without access will pay for it. Using this method, it was estimated that an annual supply of firewood equated to almost half a year's supply of maize (or 416 kg using 1996 prices) for a household, and an annual supply of poles equated to 120 days' supply of maize [25] (as cited in [13]). Using data from the National Statistical Office of Malawi, the Forestry Research Institute reports that "studies on micro-enterprises in Malawi have shown that most people who sell forest produce do so as individuals or as small family operations, start off with little, if any, capital outlay, produce small quantities of mainly unprocessed or crudely processed goods and make little profit" [13] (p. 33). As a result, little to no profit or savings are generated, as monies earned are used to meet immediate domestic needs rather than used for savings and/or investments.

On a national level, forests account for 12% of Malawi's natural capital [26]. A highly conservative underestimate is that Malawi's forests contribute 6.2% of GDP [27]. Another more comprehensive evaluation reveals that forests contribute 8% of GDP with a substantial total economic value and enormous economic contribution to livelihoods [26]. These underestimates do not take into account many environmental protection, goods, and ecosystem services provided by forests. Taking a small subset of the services provided by soil, forests, fisheries, and wildlife into account, it has been estimated that Malawi's GDP would be higher by 5.3% per year (2007 prices) were it not for unsustainable use of these resources. The discounted cost of damage over a 10-year period equated to 21.4% (in 2010) of GDP [28]. Over half (11%) of this 21.4% value was attributed to forests with only wood products, flood prevention, and air pollution services taken into account. The unsustainable use of forest resources and loss of economic value is negatively impacting Malawi's growth, where resources are limited and even small gains in natural and economic resources could have significant impacts on people's ability to meet basic needs.

#### *1.3. Response to Deforestation*

The national government recognizes the crucial regulating services (e.g., climate regulation, moderation of extreme weather, soil erosion, and pollination), habitat/supporting services (e.g., biodiversity), and cultural services (e.g., tourism) of forests [29]. Citizens are more aware of the provisioning services such as food, raw materials, and medicine that are acutely tied to their daily lives, but there is increasing awareness of the regulating services forests provide to maintain supplies of fresh water, prevent flooding, protect crops from wind damage, stabilize soil, and avoid excessive siltation of riverbeds downstream.

The largest threat to these services is occurring on customary lands, where ownership or usufruct rights (rights held by a member of the land-holding community in customary freehold [9]) "rest with individual villagers or group of individuals who the customary authority have appointed or delegated temporary ownership to, otherwise all authority rests with the customary leaders" [13] (p. 25). These lands are controlled by people residing in villages and the Traditional Authorities who oversee land tenure and land disputes on customary lands. Little research has been conducted at the local level on forests located on customary land [30]. In line with Malawi's move towards decentralization and community-based forestry, local level interventions are key to understanding country-level trends and policies that address deforestation [31].

#### *1.4. The Misuku Hills Study*

The Misuku Hills of Malawi have been recognized as one of the world's Key Biodiversity Areas [32]. There are three small forest reserves in the Misuku Hills, but it is the largest contiguous area of remaining forest in Malawi on customary land outside of public parks and forest reserves. This is most notably due to its comparatively low population. To maintain the integrity of this slowly dwindling forest, the policies that govern forest conservation must be understood for their applicability in this region and their pertinence to local control of forest resources and management of ecosystem services. There is a great need for local ecosystem service assessments in Africa that capture local resources and needs [33].

There is a long history of forestry conservation in Malawi, and the most recent forestry policies recognize deforestation as a cross-cutting issue with various drivers and impacts on a myriad of development goals. In line with the purpose of this special issue of *Forests*, we explore the policies aimed at forest protection in Malawi and discuss the challenges and opportunities faced in their implementation nationally and locally in the Misuku Hills bioregion. The research activities of this inter-disciplinary study are therefore threefold. First, an assessment of national forestry policies was conducted to determine how Malawi's forestry policies have evolved and how current policies aim to guide forestry conservation activities. Second, an inventory of the forestry conservation activities in the Misuku Hills region was conducted to determine how these policies are being applied to protect one of Malawi's last remaining indigenous forests on customary land. Third, an ecosystem services assessment was conducted using a Photovoice methodology in a remote village on customary land in the Misuku Hills.

The results reveal that Malawi has enacted a plethora of forestry policies that have been developed to meet international standards. While these policies should be lauded, the case of the Misuku Hills reveals that the full potential of these policies is underwhelming due to weak implementation and shortcomings in empowering communities to practice local control. The ecosystem services inventory conducted in Chikutu village reveals that community members have extensive knowledge of forest resources, and these resources are still readily available to the indigenous communities. These findings underscore the important role that community-based and indigenous land tenure regimes play in advancing national forest policies. The paper begins with a description of the materials and methods of analysis, followed by the results on national and local forest conservation policies and practices. The paper concludes with an examination of implications for forest management in the Misuku Hills forested region and proposals for future research.

#### **2. Materials and Methods**

#### *2.1. Malawi Forestry Policy Evolution and Analysis*

The search and retrieval of Malawi government forestry policy and guideline documents was conducted online and through personal communications with government officials, representatives of non-governmental organizations (NGO), and faculty at Mzuzu University. Policy document retrieval was further guided by a previous outline of forestry policies and by contemporary policies and guidelines that referenced previous policies. The search was focused on policies and guidelines directly related to forestry, although policies that included forestry conservation (e.g., environmental policy) were also reviewed. When the search for documents began to reveal no new information, and the body of policies could be considered comprehensive and complete, the policies were then reviewed for changes that have occurred over time. Some of the changes to previous policies were noted in the subsequent policies themselves. The evaluation of policy changes took into account changes in the form of government in Malawi and advancements and best practices in international forestry policies.

#### *2.2. Conservation Activity Inventory in the Misuku Hills Forested Region*

#### 2.2.1. The Misuku Hills Forested Region Study Area

The Misuku Hills are part of the southern end and western branch of the East African Rift System within the Southern Rift Montane bioregion [34] and the Miombo woodlands ecoregion. The Misuku Hills forest, like other remaining forests in Malawi, is in upland and hilly areas and along the rift valley scarps [7].

The Miombo woodlands of the Misuku Hills are the most floristically diverse in the country [35], encompassing the vegetative/biotic communities of open canopy woodland of hills and scarps (*Brachystegia spp.*), open canopy woodland of plateau (*Brachystegia*/*Julbernardia*/*Isoberlinia*), closed canopy woodland of wetter uplands (tall *Brachystegia spp.*), and montane evergreen forest [36]. Within the Misuku Hills, there are three public forest reserves (FR): Matipa FR (1055 ha), Mughese FR (771 ha), and Wilindi FR (937 ha), all established in 1948 [6]. These forest reserves are largely composed of montane evergreen forest. The reserve land area comprises only a small fraction of the larger contiguous forest referred to in this study as the Misuku Hills forested region (MHFR). The MHFR spans a number of political and administrative boundaries, and data from these administrative units are presented below to provide context to the MHFR study area (Figure 2).

**Figure 2.** Location of the Misuku Hills forested region (MHFR) study area.

The area commonly identified as the Misuku Hills is within the administrative territory of Traditional Authority (TA) Mwenemisuku, but, from a biotic perspective, the contiguous forest of the MHFR spans two TAs and the country borders of Malawi and Tanzania (where they are referred to as the Umalila mountains). The focus of this study is confined to MHFR in Malawi.

Malawi is divided into three administrative regions (north, central, south), and the MHFR is one the most remote areas in the northern region of Malawi, where the most dramatic and highest concentration of hills and scarps are found. Since the country's independence in 1964, there has been a gradual migration of people to the northern region [37], but the effects of this migration on development in the region have been minimal. Only 13% (2,286,960) of the total population of Malawi lives in the northern region [12]. Not surprisingly, the northern region remains by far the most forested, with 48.7% tree cover as compared to the central (26.3%) and south (27.3%) regions [5]. Compared to Malawi's two other regions, the northern region has over double the rate of those reporting use of forest resources from their own land (9.9%), the highest proportion of forest resources use from communal land (18.2%), and the lowest proportion of those buying forest products from someone else (44.7%) [38]. Table 1 depicts the proportion of use and source of forest products in these two districts [38].


**Table 1.** Proportion and source of forest resources in study area districts.

Malawi's three regions are divided into a number of districts. The MHFR spans the borders of two of the seven districts within the northern region, Chitipa and Karonga. The population density of Chitipa (54/km2) and Karonga (107/km2) are both well below the country average of 186/km<sup>2</sup> [12]. This population density estimate for Karonga exaggerates the population density in its forested areas. Karonga is topographically unique, as it is split north to south between coastal lake plains to the east and hills and scarps to the west. The overwhelming majority of the population resides in the lake plains. The population density in the hills and scarps of Karonga is similar to that of Chitipa district.

It is difficult to accurately estimate the population of the MHFR study area, as it overlaps census units, but most of the MHFR is within the administrative unit of Traditional Authority (TA) Mwenemisuku. Field observations in the area of the MHFR outside TA Mwenemisuku confirm that it is very lightly inhabited. In 2018, the population of TA Mwenemisuku was estimated at 25,816 [12]. Strangely, this is approximately 2000 fewer people than the 2008 census estimate. The land area of TAs was not available, but TA Mwenemisuku is similar in size to other TAs in Chitipa. Since its population is about average for TAs in the district, its population density is likely very close to the very low population density for the district (54/km2). A hot spot of change analysis reveals that while some districts in the northern region experienced significant forest and natural vegetation loss from 1990–2010, Chitipa and Karonga experienced no dramatic change [5].

Mapping these proportions allows us to view the MHFR as not conforming to administrative boundaries, and reveals that it has the largest contiguous area of community forest and woodlot opportunities in the country, high forest management of natural forest opportunities, and high priority for food security and biodiversity intervention [39]. This leads us to examine the interventions that have been conducted in the MHFR to pursue these opportunities and protect its forests.

#### 2.2.2. Conservation Activities Inventory

The conservation activities inventory aims to remedy the lack of accurate accounting of the past and current conservation activities in the MHFR. The search for forestry conservation activities in support of forestry policies was conducted through an online search and through personal communications with government officials, representatives of various conservation-oriented NGOs, and faculty at academic institutions in Malawi and the United States (US). The inventory of conservation activities in the Misuku Hills was further guided by a historical summary of community-based forest management activities nationwide [40]. The most insightful sources of information came from NGO project reports and personal communications with representatives of NGOs and sponsoring agencies who have operated in the Misuku Hills, as well as a current US Peace Corps Volunteer who lives and works in the Misuku Hills performing environmental conservation activities.

Since the vast majority of the land area of MHFR is on customary land and not in the montane evergreen public forest reserves, an ecosystem services assessment was conducted in the representative village on customary land to determine the extent and variety of services reaped by residents in these typically overlooked areas within the MHFR.

#### *2.3. Chikutu Village Ecosystem Services Inventory*

Chikutu village is located in the MHFR to the east of public forest reserves and just outside the eastern boundary of TA Mwenemisiku in TA Kilupula in Karonga district. TA Kilupula, like the aforementioned Karonga district in general, is split north to south between coastal lake plains and hills and scarps. Chikutu is located deep in the lightly populated hills and scarps to the west of the lake plains. It is accessible by an unimproved, and at times impassible, dirt road. In ideal conditions it takes a 4 × 4 capable vehicle or motorbike approximately two hours to scale the hills to reach Chikutu from the main M1 highway paralleling the lake shore to the east.

This area around Chikutu is less populated than the area around the forest reserves to the west and represents one of the last hideouts in Malawi, where residents could directly reap ecosystem services from largely intact natural forests. If there is anywhere in Malawi where Malawians can still reap an array of benefits from a relatively intact natural forest on customary land, this is the place. It provides a lesson for the larger MHFR, the vast majority of which is on customary land.

The unique topographical and population dichotomy of northern Karonga makes Chikutu an interesting case. Chikutu, and other villages in the area, could be facing pressure to deliver forest goods to the much more highly populated areas in the coastal plains that are almost exclusively agricultural lands with greatly diminished forest resources.

Google Maps® designates this area as the Matipa Complex Forest. It is not clear how this designation was made, as there is no discernable local knowledge of this designation or reference to it in any forestry policy or conservation plans.

#### 2.3.1. Ecosystem Services Assessment Photovoice Exercise

Photovoice is a praxis-based qualitative tool that enables participants to record, reflect, and produce knowledge on their community needs, experiences, strengths, and concerns through specified photographic techniques [41–43]. Participants use photography to represent their perspectives and lived experiences on a given topic and collectively discuss and analyze photos to inform community projects and advocate for their interests. Regarded as a tool to give agency to disempowered and marginalized groups in transforming their realities, Photovoice has been used in several disciplines, such as urban planning, education, public health, and sociology [44–49]. Because of the wide appeal of participatory photographic methods, it has been used with children, youth, and adults in various settings, ranging from youth programs, women's groups and organizations, to public health organizations [50–53].

In the context of this research project, Photovoice was used to provide deeper insight into the context of rural resource dependent communities and the ways in which groups derive value from their relationships with forest ecosystems. Few studies on ecosystem services assessments have utilized Photovoice as a research method [54,55], which makes this work unique and fills a significant gap in the literature.

Prior to conducting the Photovoice ecosystem services assessment, a number of necessary permissions were obtained. The Principal Investigator (PI) applied for and was granted the necessary permission to conduct research in Malawi from the National Committee on Research in the Social Sciences and Humanities under the Malawi National Commission for Science and Technology (Ref No: NCST/RTT/2/6). Once on site, the PI and co-PI met with TA Kilupula to inform him of our activities and seek his permission, which was granted. TA Kilupula oversees 16 Group Village Headmen (GVH), and hundreds of Village Headmen/women in northern Karonga, including the GVH in Chikutu. The PI and co-PI also informed the GVH in Chikutu of our arrival date and research prior to the data collection visit.

The co-PI is a Malawian who has lived in northern Karonga his entire life. He is an extremely well-respected member of the community, is fluent in the handful of languages spoken in the region, and is an employee of the Ministry of Health of Malawi. The PI has known the co-PI for nearly 20 years, two of which (1998–2000) were spent working together at a rural hospital in northern Karonga. Prior to fieldwork, the PI and co-PI met many times in Malawi to discuss the purpose of the study and to refine the Photovoice methods and procedures.

Photovoice involved community members photographing the lands and natural elements around their villages that they perceived as essential to their health and livelihood. The exercise replicated the Misuku Hills biodiversity and livelihood transect walk conducted in 2015 with Village Natural Resource Management Committee (VNRMC) members in the area around the forest reserves [56], but added photographs of ecosystem services identified by residents. The assessment was conducted throughout the mid–late summer months when many resources are in season and available, but data on resources not available during these months were also collected.

The Photovoice methodology was applied to the ecosystem services assessment to overcome the limitations in articulation with text-based research and the asymmetrical power balance inherent in other research techniques. While allowing research participants to drive the process might frustrate the answering of a narrow set of questions and evidence is often not generalizable, it is abundantly useful for "building participant-driven practical theory about how environments impact everyday people" [49] (p. 400). The Photovoice variation used in this study combining a walking tour with picture taking replicates the method used in an informal settlement in Lusaka, Zambia [49]. This method gives a significant level of control over to participants and roots the data in lived experiences.

Upon arriving in Chikutu, we began by meeting with the Village Headman and the study participants to explain the purpose of the visit and to receive informed consent. The Photovoice participants included four men and three women of ages ranging from adolescent to adult. The field research in Chikutu proved to be an opportunity to educate residents on informed consent standards and procedures. Before the Photovoice exercise, the participants, PI, and co-PI reviewed the informed consent form that had been translated into the local language (Chitumbuka). One representative from the participants read aloud each section to ensure that any illiterate participants understood what was being consented to, and, following each section being read aloud, the co-PI reiterated the information and asked if any participants had questions. Although Chikutu is very remote, residents have had occasional contact with government officials collecting data, but none of the residents had ever completed an informed consent. After the form was signed, the PI and co-PI encouraged residents to demand informed consent from future researchers so that they would be fully aware of the risks and benefits of participating in research.

Once the Photovoice ecosystem services inventory walk commenced, the participants pointed out a resource, a photograph was taken of the resource, and its local name and common uses were recorded. The participants were given slight prompts to provide a little more explanation when necessary, but at no point did the researchers independently point out a potential resource and ask "what is this?" The participants were given full control to point out the resources that were important to them for their health and livelihood and to end the exercise when they felt a full accounting had been collected.

#### **3. Results**

#### *3.1. Malawi Forestry Policy*

The data from this section are aimed at addressing the first of the three research questions: What are the forestry policy challenges and opportunities for forest conservation in Malawi?

Timeline of Malawi forestry policies and guidelines and selected other policies directly affecting forestry policies:


There has been a clear evolution in forest policies under Malawi's three distinct forms of government that span its colonial demarcation as a country to its current multi-party democracy system. During the colonial era's command and control system of management (1890s to 1964), "forest guards were posted in every Traditional Authority ... to police forestry use and collect revenue for government from forestry products" [7] (p. 2). With independence from colonial rule in 1964, and under a new authoritarian one-party state, the focus turned to plantation timber production for local and international trade. Concurrently, forested land under the control of traditional leaders experienced accelerated deforestation and degradation as communities pursued extractive practices as a demonstration of political independence from a colonial system of forest management [7].

During Malawi's transition to a multi-party democracy in 1994, demand for forest goods and services far exceeded supply, putting further pressure on forest systems [58]. To address recognized environmental and forest issues in Malawi, the Forestry Policy was revised in 1996. The 1996 National Forest Policy was a departure from the traditional forest management approach, most notably with its marked move towards devolution of centralized powers to promote participatory management [57,59,60]. This new strategy emphasized "multi-stakeholder participation including local communities" [7] (p. 2). Unfortunately, democracy was also equated with deregulation and the deforestation of forest reserves for agriculture and fuelwood [61].

Many of the devolution objectives acknowledged forestry financing and enforcement challenges. To remedy these challenges, a market approach was adopted to provide economic incentives that promoted the sustainable utilization of forest resources by emphasizing local ownership and management of forests and small- and medium-scale forest-based industries. Local management of forest resources was designed to be achieved through community-based forest management practices embedded in traditional institutions and giving communities shared or exclusive decision-making rights [7]. Another notable change in the 1996 policy was the explicit recognition that forest conservation policy objectives were supporting quality of life measures for rural populations recognized as the most disadvantaged group in Malawian society [58]. The 1996 forest policy was also used to support the larger framework of the 2004 National Environmental Policy, which itself is aligned with Section 13(d) of the 1995 Constitution of Malawi outlining many environmental goals.

Subsequent enabling legislation, forestry policy updates, and participatory management guidelines all support the current 2016 National Forest Policy goal to provide "guidance to the management of forests, offer an enabling framework for all stakeholders to participate in the management of forests, and sustain the contribution of the national forest resources for the upliftment of the quality of life" [27] (Foreward). Although the recognized role of forests in supporting quality of life is still an objective of the 2016 National Forest Policy, its focus on the rural disadvantaged is no longer explicit. While the 1996 policy was emblematic of a new democracy facing a recognized threat, the 2016 policy is more outward looking and reflective of a young participatory democracy on a world stage. This is evident in its stated alignment with international agreements and conventions such as the

Rio Declaration, United Nations Framework Convention on Climate Change, the Montreal Protocol, United Nations Convention to Combat Desertification, United Nations Convention on Biological Diversity, United Nations Convention on International Trade in Endangered Species of wild fauna and flora, and the claim that the review of the policy was conducted by a wide range of stakeholders, including traditional authorities, district councils, the civil society, the private sector, statutory bodies, government departments, academia, and the general public [27].

The 2016 policy recognizes that forestry conservation is a cross-cutting issue which requires collaboration and broad participation to meet the goals of other focused policies, such as those addressing land, biodiversity, wildlife, water, energy, and population, but also the more comprehensive Malawi Growth and Development Strategy now in its third iteration. Among the policy outcomes aimed at protecting forests are financial benefits and other livelihood outcomes (e.g., food, biomass, shelter, health). Financial incentives to protect forests include eco-tourism and recreation, and also still include forest-based enterprises. The livelihood outcomes are realized in the goods residents reap on a daily basis and profit from to support their health and well-being.

The implementation and enforcement of these policies remains a significant challenge, but the Environment Management Act of 2017 aims to address these challenges. This act created an Environmental Protection Agency "with broad and substantial powers to strengthen environmental planning and risk management at national and decentralized levels" [62] (p. 2). If the same implementation challenges that have thwarted previous legislation can be overcome, this act "will be one of the most powerful legal instruments for environmental management introduced so far in Africa" [62] (p. 2). This act, like others, focuses heavily on local control of environmental resources.

#### 3.1.1. Village Natural Resource Management Committees and Village Forest Areas

Locally developed and enforced customary laws have shown to have a greater impact on the protection of natural resources as compared to federally developed and enforced laws [63]. This is partially due to human and institutional resource constraints that continue to make government sponsored patrolling, enforcement, and prosecution a challenge. Customary control alleviates these resource constraints and aligns with the customs and rules that govern everyday life and natural resource management and sanctions in Malawi. "Locally developed and enforced resource-use rules which relate directly to the resource in question", are more easily monitored by other community members, and acknowledge a culture where subjective norms are well known within communities, weigh heavily on decision making, and traditional penalties are generally accepted [63] (p. 93).

Recognizing the necessity of local control and influence of customary law, key among the strategies to achieve the objectives of national forest policies is the establishment and support of Village Natural Resources Management Committees (VNRMC). These nationally registered committees receive technical advice from the Forestry Department officers on how to protect, control, and manage their forest resources [13]. Under Forest Rules 2001, the VNRMC has the authority to prohibit residing in protected areas, altering for agriculture, or damaging trees for any purpose (along with selected other powers). The local Forest Management Agreement created by VNRMC in consultation with Forestry Department officers governs the activities of demarcated Village Forest Areas [63].

Village Forest Areas (VFA) are areas on customary land that are actively managed by the VNRMC for forest resources or forest re-establishment. As with all customary land, and following customary law, the responsibility for allocating and overseeing the VFA lies with the traditional leadership of the Village Head, Group Village Head, or ultimately, the Traditional Authority.

During the colonial era, every village was required by law to have a VFA to oversee the conservation of wood products, water, biodiversity, and recreational facilities. "A total of 69,000 hectares of VFAs were set aside by 1940, under the control of local headmen and for the purpose of local use" [6] (p. 1). VFAs are no longer required by law, but with devolution to local level control, these VFAs remain critical to forest conservation and are often the only body overseeing forest conservation.

There were over 2000 "active" and another 1000 "trained" VNRMCs in Malawi in 2002 [6], and the scope of their activities have expanded in line with heightened knowledge of the critical role of forests in local and global ecosystems. Many VNRMCs are now tasked with reforestation, and tree planting activities often receive wide media coverage. Their influence has waxed and waned since the colonial era establishment of VFAs, but the move towards decentralization since 2008 provides guarded hope that their influence will return. Granted, it will likely take time to overcome decades of centralized control and the re-adoption of local responsibility. It is a promising sign that the large youth populations in Malawi are taking center stage as stakeholders in VFAs [64], although the national youth tree planting program has also experienced shortcomings in government administration [65].

#### *3.2. Conservation Activity Inventory in the Misuku Hills Region*

2013–2015 Promotion of Indigenous Forests in the

The data from this section are aimed at addressing the second of the three research questions: What conservation activities are taking place in the Misuku Hills in support of these policies?

It has been documented that forest management activities in Malawi often involve three parties: "a facilitator who catalyses the process (often coming from outside the community), the implementing agency (a local group or committee spearheading the change process) and the benefiting community" [7] (p. 6), and this holds true for the activities that have taken place in the Misuku Hills since 1999. The facilitators have often been Malawian NGOs backed by international aid organizations (Table 2). It is apparent in Table 2 that NGOs, and not government, have been the catalyst of forest conservation activities, with government officials acting in an advisory capacity. NGOs that operate in Malawi must be registered with the Registrar General in the Department of Justice and the NGO Board of Malawi. They may also voluntarily become a member of the Council of Non-Governmental Organizations in Malawi (CONGOMA).


Misuku Hills Area MBA <sup>6</sup> UNDP GEF <sup>8</sup>

Development Partners

**Table 2.** Inventory of sponsored forestry conservation activities in the Misuku Hills.

<sup>1</sup> United States Agency for International Development; <sup>2</sup> LTS historically stood for Land and Timber Services, but it is now just LTS as a stand-alone title; <sup>3</sup> Gesellschaft für Organisation, Planung und Ausbildung; <sup>4</sup> Critical Ecosystem Partnership Fund; <sup>5</sup> Action for Environmental Sustainability; <sup>6</sup> Misuku Hills Beekeepers Association; <sup>7</sup> During the time of the project, the organization was called Sustainable Rural Growth and Development Initiative (SRGDI), but the organization is now called Sustainable Development Initiative (SDI); <sup>8</sup> United Nations Development Program Global Environment Facility. COMPASS, Community Partnerships for Sustainable Resources Management; IFMSLP, Improved Forestry Management for Sustainable Livelihoods Program.

2017 Misuku Hills Art Challenge AfES 5, MBA 6, SDI <sup>7</sup> CEPF <sup>4</sup> 2018 Small Producers Development Project MBA 6, SDI <sup>7</sup> IM-Swedish

The remoteness of the Misuku Hills contributes to it being one of the last contiguously forested regions of Malawi not in a national park, but this feature has also caused it to be overlooked by conservation organizations and funding agencies operating in other regions of Malawi. Despite this, there have been a handful of forest conservation activities that have engaged the VNRMCs in the Misuku Hills charged with managing both VFAs and the three public forest reserves.

The two phases of the Community Partnerships for Sustainable Resources Management (COMPASS) project were national in scope with Misuku Hills as one among many intervention sites. The very limited documentation available on the activities and results of COMPASS I reveal that it was meant to improve natural resource management by emphasizing income generation, which is consistent with and supporting USAID/Malawi's Strategic Objective framework of sustainable increases in rural incomes [40]. Unlike COMPASS I, there is abundant documentation of COMPASS II [66]. COMPASS II "supported decentralized environmental management and capacity building in enterprise development in order to mainstream CBNRM [community based natural resource management] as a viable rural development strategy (COMPASS II Project, 2007)" [40] (p. 24). COMPASS activities in the Misuku Hills included the support of the Mzuzu Coffee Planters Cooperative Union (MCPCU) to engage in honey production. Honey production was promoted not only for the sale of honey and beeswax products, but also for pollination of coffee crops. The accomplishment of COMPASS targets were initially monitored across 15 districts in Malawi, including Chitipa, but then reduced funding in 2007 led to only seven districts being monitored, excluding Chitipa [67]. Therefore, the effectiveness of these programs cannot be evaluated, and this lack of evaluation proves to be a theme throughout subsequent forest conservation activities.

The Improved Forestry Management for Sustainable Livelihoods Program (IFMSLP) was a Government of Malawi, two-phase, national capacity-building exercise aimed at improving "the livelihoods of forest dependent communities through improved sustainable collaborative management of forests both in forest reserves and customary land" [68] (LTS webpage). IFMSLP I and II were aimed at the implementation of the National Forestry Policy and Program through community mobilization, institution building, and local forest management planning [69,70]. One of the IFMSLP intervention sites included the area within and around the three public forest reserves in the Misuku Hills [6]. IFMSLP II faced funding and implementation delays, but it was eventually pushed forward with a reduced set of strategies [70]. It was discovered through personal communications that the delays were due to suspected government corruption which caused the funding agency to halt the project, but it did eventually resume with "competitive grants for non-state actors to enhance their role in, and to accelerate, project implementation" [70] (p. 1924).

There is a conspicuous dearth of documentation available on the activities and outcomes of the two phases of the IFMSLP, but one available document was an evaluation of IFMSLP I completed in 2011 [71]. The evaluation was critical of the lack of monitoring and evaluation, which may help explain the lack of available documentation. Despite the program setbacks and the lack of data on program activities and outcomes, there were still some identifiable results.

On a national level, the IFMSLP led to the development of the national 2013 Standards and Guidelines for Participatory Forestry in Malawi [70], as referenced in Section 3.1.1. The IFMSLP activities and outcomes specific to the Misuku Hills intervention site were uncovered through communications with NGOs who received the "competitive grants for non-state actors" to pick up the pieces from the mixed successes of IFMSLP I and a halted IFMSLP II. One tangible outcome was the creation of the Matipa Forest Management Plan (in draft form and still under review by the Forestry Department) created under the Improved Livelihood and Biodiversity Conservation Project. This plan was designed to govern the activities of the VNRMCs in this area and create a greater sense of ownership by the surrounding communities to relieve unsanctioned pressures on the forest reserve resources. It was based on and serves the larger Strategic Forest Area Plan (SFAP) for the Matipa, Mughese, and Wilindi forest reserves, with a focus on three of the five priority objectives in the SFAP. These are to (1) increase in tree planting and natural regeneration, (2) conserve the forest for water catchment protection, unique biodiversity, and cultural heritage, and (3) regularize extraction of forest resource and products from forest reserves to uplift communities' livelihoods. The creation of forest management plans for all three forest reserves was originally the responsibility of the Department of Forestry in IFMSLP I, but only one of these plans has been prepared. The Matipa Forest Management Plan includes estimates of the products and income that could be generated from the sustainable use of forest resources and a monitoring plan for patrols by local residents [72]. Responsibility for the creation of plans for the two remaining forest reserves (Mughese and Wilindi) remained with the Department of Forestry in IFMSLP II, but these plans have yet to be prepared.

The IFMSLP II also supported the creation of a number of forest management agreements and beekeeping activities under the Misuku Hills Indigenous Forest Project [73]. Seven VNRMCs composed of members from 71 villages now have co-management agreements and licenses to oversee the entire 2762 hectares of the three forest reserves in the Misuku Hills. These agreements include tree planting and regeneration on 485.92 ha in the forest reserves. These VNRMCs simultaneously manage VFAs on customary land, and there were seven VFAs established under the program (Alther, Chipala, Chiwi, Kapiyira, Lupalang'ombe, Mwenga, Nangalamu) with four of these VFAs (Alther, Chiwi, Mwenga, Nangalamu) currently with completed VFA management plans. Although there is no readily available national accounting of VFAs or their level of current activity, a forestry officer from Chitipa reported that there are 68 VFAs around the Misuku Hills, and 21 of these VFAs have management plans.

The Misuku Hills Indigenous Forest Project also included activities in and around forest reserves focused on beekeeping, candle making, and selling non-wood products. The Misuku Beekeepers Association (MBA) was the lead organization in this effort and has proven to be a prominent force for forest conservation in the Misuku Hills. MBA is a registered company comprised of more than 1500 beekeeper members with a 2:1 ratio of men to women [74]. MBA has actively contributed to the creation of forest plans, and their forest-based enterprises were further enhanced through this project. Support from this project led to MBA being chosen to represent Chitipa district at the National Agricultural Trade Fair, where they established business connections and received beekeeping product orders. These activities were estimated to have increased the income of 350 households in the area by 80%. MBA members also increased capacity in project management skills and conservation science by participating in two workshops in Mbeya, Tanzania and Nairobi, Kenya. Their participation at these workshops led the MBA being nominated by the Tropical Biology Association for a site visit and learning exchange with Save Tanzania Forests.

The Misuku Beekeeping Value Addition Project (MBVAP) and Promotion of Indigenous Forests in the Misuku Hills Area (PIFMH) project occurred simultaneously with the Indigenous Forest Project discussed above. The MBVAP was aimed at continuing to build capacity in the forest-based enterprise of beekeeping among the estimated 2500 beekeepers in the Misuku Hills who produce honey at a subsistence level. Honey was being sold unprocessed and uncertified in unreliable markets in their area, and the beeswax was just thrown away. The project supported the training of 250 members of 50 beekeeping clubs in beekeeping techniques and honey and beeswax candle making; the provision of equipment and materials; certification of honey and candles; and linking farmers to markets [74]. The reported results of the project were "overwhelming." The 50 clubs that were trained were registered with MBA. MBA acquired certification from the Malawi Bureau of Standards and used the processing equipment to create professional and standardized packaging for their Misuku Hills Honey (Figure 3). The combination of professional packaging and certification was attributed with over an 80% increase in the value of their honey. MBA honey can now be found on grocery store shelves across the country, and it is being used as an ingredient in cough syrup by a pharmaceutical company. This has reportedly led to 300 households increasing their income by 50%. Although the wax processing and candle making plant was established and most of the clubs 50 clubs were trained, inadequate funding was given as a reason why the candle making venture has yet to be realized.

The PIFMH project also promoted beekeeping, but it was focused solely on the Mughese forest reserve and included the promotion of nutritional and medicinal products and "reconnecting cultural values with existing nature" [75] (webpage). Although no evaluation report on outcomes of the PIFMH was available, its status is listed as "satisfactorily completed" [75].

The relationships built between stakeholders and organizations in all of these projects led to further collaboration in the Misuku Hills Art Challenge (Figure 4) aimed at raising "awareness of the beauty and ecological, cultural, aesthetical and economic value of Misuku Hills Forest Reserve both locally and internationally" [76] (webpage).

**Figure 3.** Misku Hills Honey. Source: Sustainable Development Initiative.

**Figure 4.** Misuku Hills Art Challenge: (**a**) Beekeeping training; (**b**) painting submission; (**c**) prize winning sculpture. Source: Sustainable Development Initiative.

The Misuku Hills Art Challenge (MiHAC) was a national competition that brought together 12 Malawian artists, photographers, and film makers to showcase the remote and often overlooked Misuku Hills and bring greater national and international attention to this Key Biodiversity Area (KBA) threatened by logging, charcoal production, and agricultural expansion. A smaller art competition was also conducted for school children in 10 schools surrounding the Misuku Hills. MiHAC was widely advertised through the national media and on multiple social media platforms, and an exhibition of the 16 paintings, 3 sculptures, 68 photographs, 3 films, and 10 children's drawings was convened in the capital city of Lilongwe with cash prizes given to the top artists.

The impacts of the project not only raised awareness of the Misuku Hills among the public, but also engaged numerous government ministries to encourage the inclusion of the Misuku Hills into policies and plans as a KBA. The Misuku Hills is recognized internationally as a KBA [32], but it is not currently recognized by the Government of Malawi as such. Participation by the Environment Affairs Department (Biodiversity Focal Point) brought attention to the need to conduct a biodiversity assessment as the first step in it being recognized as a KBA in national policies and plans.

MiHAC also introduced tourism as a new approach to forest conservation in the Misuku Hills, and tourism brought in the Ministry of Tourism as a new player in addition to forestry. The only accounting of tourism in the area estimated that the Misuku Hills typically received three local and international tourists per month, but in the six-month period following MiHAC, this rose to eight tourists per month, nearly tripling the income of VNRMCs that charged small fees to tourists [77]. The MiHAC project also brought greater attention to the production of Misuku Hills Honey and Mzuzu Coffee as potential tourist attractions. In addition to the previously discussed success of marketing Misuku Hills honey, the Misuku Hills is also where 50% of the internationally-renowned Mzuzu Coffee is produced. A small number of tourists travel to the Misuku Hills to sample the coffee and learn about the community that grows the coffee.

A number of potential donor organizations were also invited to and attended MiHAC events, and the Small Producers Development Project resulted from the interactions with a donor. A pilot project was conducted to continue the work of supporting beekeepers with financing, certification, production, and marketing, but it was reported that the donor ultimately decided that the Misuku Hills were too remote to conduct monitoring and evaluation, so they ceased supporting beekeeping activities. Again, the double-edged sword of its remoteness being part of its attraction but also creating difficulties in accessibility.

Lastly, there appears to have been a small project funded by the Tilitonse Foundation aimed at strengthening VFA plans, but no documentation on this project was available.

Although not in direct administration of the conservation activities in the Misuku Hills listed above, the Wildlife and Environmental Society of Malawi deserves mention as a major player in conservation activities in the Misuku Hills and throughout Malawi. They were credited as a contributor to many of the activities and policies identified in this study.

#### *3.3. Chikutu Village Ecosystem Services Inventory*

The data from this section are aimed at addressing the final research question: What ecosystem services are residents using that are in need of protection?

The Photovoice ecosystem services inventory took approximately three hours to complete. The residents of Chikutu village identified 16 distinct forest products in the immediate vicinity around their village (Table 3). This included seven different types of fruits, roots used for medicine, fiber and timber used for construction, vegetation that prevented soil erosion, the soil itself, mushrooms, and grasses and leaves for domestic animal feed. Figure 5 displays selected photos of identified forest products.

A greater volume and variety of flora were identified in the Chikutu ecosystem service Photovoice exercise than were identified in a previous participatory assessment conducted as part of the Improved Livelihood and Biodiversity Conservation Project.

Forest fruits are an essential part of the diet for rural populations in Malawi as they are a source of critical dietary nutrients (vitamin A and C, calcium, fiber, minerals) and contribute to food security, especially as a supplement in times of famine. Mushrooms also provide nutritional benefits and are sometimes preserved for food security purposes [25]. Fruits and mushrooms can be sold to persons residing in surrounding areas where forests have been degraded and these resources are scarce [13].


**Table 3.** Inventory of forest products.

<sup>1</sup> Almost all trees are also used for wood fuel.

**Figure 5.** Selected forest goods in Chikutu: (**a**) Ndilolo (cashew-like nut); (**b**) Chiwowa (mushroom); (**c**) Munyere (wild avocado).

The remoteness and inaccessibility of Chikutu village has contributed to the abundance of these forest products, but lack of connectivity to other settlements also creates great difficulty in accessing markets. This lack of access to markets makes these goods essential as a food source and, at the same time, creates great difficulty in selling these products for profit. When asked directly about the sale of forest products, it was confirmed that they were only consumed locally. One notable absence in the identified forest products was the harvesting of wild animals and insects.

A number of ecosystem service assessments were also conducted in three villages in the lake plains area of Karonga with much higher population densities and diminished forests. While residents in these villages still appeared to be knowledgeable about forest product ecosystem services, these services were less abundant and diverse. Essentially, the less forested area had fewer identified ecosystem services.

#### **4. Discussion**

Malawi is faced with many challenges to forestry conservation and the subsequent sustainability of ecosystem services. However, as the results above revealed, Malawi has formulated—in collaboration with an array of international organizations—a number of policies and plans to meet these challenges. There have been attempts to implement these policies, but there are significant challenges to realizing their full vision and potential. The first challenge is a lack of current, reliable, and comparable data on forest cover and change, but efforts to remedy this are currently underway (Appendix A). Discussed below are a number of other challenges, as well as potential remedies discovered through this research: The challenge of good governance, the potential remedies of increased location control (e.g., Village Natural Resource Management Committees), and the promotion of forest-based enterprises. As the results of the policy inventory revealed, there is no shortage of actionable forest policies, but shortcomings in good governance has limited their implementation. The forest-based activities that have taken place in the Misuku Hills demonstrate the potential for local initiatives and control when centralized government fails. The inventory of ecosystem services revealed a previously unrecorded wealth of goods that could be sustainably marketed through forest-based enterprises. The discussion concludes with suggested directions for future research.

#### *4.1. Good Governance*

Weak institutions have been identified as one of the many factors threatening the Miombo forests of Africa [78,79]. The Democracy Index categorizes the relatively young multi-party democracy of Malawi as a *hybrid regime* [80]. The hybrid regime-type of government is characterized by a number of conditions that affect the implementation of forest policy, including serious weaknesses in political culture, functioning of government, and political participation; widespread corruption; weak civil society and rule of law; and a non-independent judiciary. Many of these threats to good governance were confirmed in a separate, albeit ideologically-driven, analysis of economic freedom [81]. Combined with a lack of financial and human resources, these conditions certainly create challenges in the implementation and enforcement of forestry policy, and they are noted as a priority focus area in the most recent National Forestry Policy [27].

We see some of these challenges at the local level in the Misuku Hills. Among the reported lessons from one of the conservation projects in the Misuku Hills (intentionally not identified) was that corruption is a key challenge. It was reported, but not corroborated by the author, that government authorities accepted bribes to grant logging licenses to traders to cut down large trees in VNRMC co-managed forests without the consent of the VNRMC. These activities undercut the stewardship and disenfranchised the VNRMCs. Part of this can be explained by the lack of internalization of the co-management concept by government officials, despite the abundance of adopted national policies and guidelines promoting participatory forest management [82]. Throughout Malawi, residents and VNRMC members have demonstrated that they are willing to put in the effort, but they are bogged down by lack of support from the government officials with whom they co-manage forests. Even when drafts of local Forest Management Plans are created, there are significant delays in approving these plans, which demoralizes residents [82]. The Misuku Hills are a case in point. There has been a many years long delay by the Department of Forestry in approving the draft Matipa Forest Plan and finalizing the plans of the other two forest reserves, long after the local VNRMCs have done their part.

Despite these challenges, there are a great number of opportunities in good governance in Malawi. The national government has delivered on forestry policy, forest management acts and guidelines, and most recently a strategy for measuring progress on forest conservation. Furthermore, the total land area of public protected forests increased by 8% between 1998–2010 with more proposals for protected areas underway [29]. Malawi has also taken the bold step of temporarily deploying Malawi Defense Force soldiers to patrol the most threatened public forests [83]. Furthermore, it was reported that the National Tree Planting Season recently closed with 50 million trees planted on 25,000 hectares, just shy of the 60 million tree target [84].

Despite some negative and demoralizing interactions between VNRMCs and government officials, there have been positive interactions as well. The Misuku Beekeeping Value Addition Project reported "a lot of support" from the Ministry of Agriculture and Food Security, Community Development Officers, and Forestry Officers in the form of training and equipment and identifying markets [74]. The Forestry Department has served in advisory and administrative roles (see Indigenous Forest Project report), and their role will likely become more advisory with the continued movement towards central devolution to more local control.

#### *4.2. Local Control*

Malawi adopted a participatory forest management policy in 2001 (updated in 2003) and, as recently as 2013, adopted guidelines for participatory forest management to increase local control of forest conservation. There is ample guidance and authority for local control, but co-management challenges are still thwarted by a lack of policy implementation and capacity building at the local level [62,70]. This lack of implementation begins with government ministries and donors failing to fully engage and organize communities. Even in communities where VNRMCs have been organized, there have been many other challenges to their effectiveness in advancing local control. Two of these challenges are (1) a lack of empowerment to actively participate in decision-making and (2) a lack of downward accountability among leaders which has limited the devolution process [85].

Much of the community-based forestry management (CBFM) has been driven by donors, government, or NGOs and imposed on communities. While these activities may raise community awareness, they are not sustainable [70] and "undermine achievement of conservation and social goals" [86] (p. 687). They lack the community empowerment derived by involving community members in decision making and creating accountability mechanisms. COMPASS I and II are examples of imposed programs that do not properly empower communities or provide sufficient incentives for communities to continue the imposed project once the donors move on and funding ceases.

Empowerment in the decision-making process begins by understanding "the preexisting conditions and how communities understand and interpret the program" [87] (p. 338). One of these preexisting conditions includes the power relations in communities. In an evaluation of CBFM in southern Malawi, it was found that both the CNFM (community natural resource management) concept and implementation created new elites (forest committees) who largely operated as corrupt, unaccountable "village bureaucracies," alienating communities from CNRM. Widespread forest degradation and institutional breakdown ensued. Community management became committee management, and part of the problem. Rare "success" was associated with idiosyncratic leadership qualities of village heads, suggesting need for enhancing roles and leadership skills of traditional leaders in balancing the exercise of power among CNRM stakeholders, and for broad-based community empowerment so that members can demand accountability from local leaders [86] (p. 687).

Both the CBFM committee and community members must understand that the CBFM committee is working for, and accountable to, the community.

Despite these challenges, there are a great number of opportunities for local control and the local protection of forests in Malawi. In fact, with a dearth of government resources to manage and monitor forests effectively, local control is currently Malawi's only hope for forest conservation. It has been found that VFAs under participatory forest management (PFM) had higher tree species abundance and diversity than those without PFM [88]. This success was attributed to the regulation of access and the forest development work of communities who practice PFM in their designated VFAs.

As noted above, understanding the preexisting conditions is essential for the empowerment that would enhance local control. A fundamental preexisting condition that is ripe for success in its ability to engage communities and build understanding and trust is undertaking a participatory process in the inventory of local forest resources, such as the ecosystem services (ES) assessment conducted in this study in Chikutu. "Considerable indigenous knowledge and skills for managing forest goods and services are often available at village level" [16] (p. 6), and it is arguably remiss to protect forests without a locally-driven accounting of the resources that are available and in need of protection. These activities are in line with current policy. The Chikutu ES inventory supports the policy priority area five, strategy three of the National Forestry Policy noting indigenous knowledge acquisition and dissemination [27]. The Chikutu ES study found a number of services that were distinct from the more heavily studied area immediately surrounding the public forest reserves. It is only by engaging communities and respecting their knowledge to uncover these services that management priorities can be properly represented in local VFA plans.

#### *4.3. Forest-Based Enterprises*

Forest-Based Enterprises (FBEs)—interventions meant to empower communities to realize tangible benefits from forests—recognize that forest conservation and livelihood development are deeply intertwined [89,90]. FBEs tie people to local forests for their livelihoods, and this may be more realistic in the near-term as Malawi continues to defy the conventional development pathway and macro-level economic changes theorized as affecting forest conservation.

Usually, the transition from a low- to middle-income economy starts with an abundance of natural capital which is used to invest in infrastructure (produced capital) and education and health (human capital). At middle-income levels, produced capital roughly doubles its share and human capital grows rapidly to become the main asset. In Malawi, the opposite development occurred. Malawi is still highly dependent on its natural capital, which remained constant at 43% from 1995 to 2014, while human capital increased only slightly and produced capital shrank [62] (p. 5).

With human capital and produced capital near stagnation, Malawi has yet to follow the forest transition theory [91], which posits that economic development, industrialization, and urbanization causes an initial large decline in forest cover to fuel this development, followed by a slow increase in forest cover [92,93]. "In some places economic development has created enough non-farm jobs to pull farmers off of the land, thereby inducing the spontaneous regeneration of forests in old fields" [92]

(p. 23), but Malawi is still a largely agrarian-based economy, and FBEs may be asking people to be tied even more to the land instead of generating economic activity off of it, possibly perpetuating a "poverty trap" [92] when FBEs are not successful.

As Malawi continues to defy the conventional development pathway and the conditions necessary for a forest transition, FBEs may be a strategy only in areas where forests are still present. In the central and southern regions of Malawi, where the vast majority of the population lives, the near complete deforestation does not make FBEs a widespread solution to poverty alleviation and forest stewardship. The northern region (and the Misuku Hills, in particular), due to its lighter population and accompanying extant forest cover, is likely a better candidate for viable FBEs. Yet, FBEs may only be viable if the conditions that led to the deforestation of the central and southern regions are addressed; namely, a rapidly growing population of small landholders practicing largely subsistence agriculture on finite land. It is not until other developments that create technological separation, reduced population, urbanization, or all of these and other factors ensue before the people of the Misuku Hills can make FBEs more than a short-term solution. If larger external forces continue to diminish forests, there will be no resources remaining to sustain FBEs.

It is also critical to recognize that "households will not invest precious labour and time nurturing trees when there are more pressing needs for food security" [16] (p. 6). Sustainable agricultural practices must be implemented concurrently with the CBFM of "free" food stuffs from forests to ensure food security as a necessary precursor to the protection of forest resources.

There has been some successes of FBE in other regions of Malawi to protect the few remaining forests (e.g., Sustainable Management of Indigenous Forests program in Kam'mwamba, Neno District in southern Malawi, 1996–2006 (see [7])), and also in the Misuku Hills with honey production, but the remote location of the Misuku Hills is a proverbial double-edged sword as it attributes to the quality of its forest but also creates challenges in transporting goods to markets and attracting all but the most adventurous tourists. One of the goals of the Misuku Hills Art Challenge was to raise awareness of the beauty and potential of this region for tourism. Even a slight increase in tourism to this area could have a dramatic effect on increasing recognition and spurring greater infrastructure development.

There has also been some success in the changing attitudes towards the protection of goods used to support FBEs. With awareness raised as to the importance of forest conservation to FBEs, there have been community protests and bad press when trees are harvested. For example, the District Forestry Officer overseeing the Misuku Hills had to address outrage at tree harvesting in the press, even when trees were being harvested sustainably on customary land and the trees in question were planted by inhabitants [94]. This culture of protection does have its limits. For example, although charcoal production is illegal and has been widely recognized as a major contributor to deforestation in Malawi, there are many instances of persons selling charcoal in plain view of those responsible for enforcing this infringement. This burgeoning culture of conservation may be driven by not only the potential losses of cash value, but also by the highly understudied non-cash value of forests.

#### Non-Cash Value

Zulu 2013 offers *reciprocal altruism* as a meta-theory [95,96] to understand the non-cash incentives for forest conservation in Malawi. This theory involves the "the trading of altruistic acts in which benefit is larger than cost so that over a period of time both enjoy a net gain" [70] (p. 361). Since there are many conditions that do not make FBEs viable in most places in Malawi, non-cash benefits may be more appropriate to remedy the "generally poor results" stemming from the "heavy dependence" on cash incentives in contemporary co-management projects [70].

In his critique of the national IFMSLP (see Table 2), but also applicable to the many programs carried out in the Misuku Hills, we tend to agree with Zulu that "a narrow emphasis on cash incentives as the motivation for 'self-interested' users to participate in co-management overlooks locally significant non-cash motivations, inflates local expectations beyond ability to deliver, and often creates perverse incentives that undermine socio-ecological goals" [70] (p. 1919). The promise of cash incentives in FBEs have been found to produce "modest early gains in institutions and capacity building and forest condition, but low and generally disappointing cash benefits." The disappointing results "burdened poor communities with conservation costs and created perverse incentives to overharvest, be dependent on the project/government, and to marginalize the local poor" [70] (p. 1919).

An additional assessment of CBFM in Malawi confirmed that many communities "also highly valued other non-cash based and environmental objectives and benefits, including the sustained access to firewood and NTFPs [non-timber forest products], social/religious services inside the forest, and continued water supply for consumption and irrigation" [82] (p. vii). Awareness of the non-cash value of forest ecosystem services appears to be high and might help explain a continued commitment to CBFM in forest reserves despite FBEs producing "disappointingly low financial benefits for poor communities burdened with conservation costs" [70] (p. 1936). In the ecosystem services assessment of Chikutu, no cash values were expressed, even when residents were explicitly asked about the potential to sell local forest products.

#### *4.4. Future Research*

This study is the first inventory and analysis of the policy potential, conservation activities, and ecosystem services in the Misuku Hills. This initial assessment aims to enable future studies to examine the impact of these policies and activities directed at protecting forests and the essential ecosystem services that forests provide to the indigenous populations in this region.

The primary suggestion on future research design is that the geographical scope of the Misuku Hills should be expanded to encompass the forested area on customary land. The larger Misuku Hills forested region on customary land has been completely ignored in research and largely ignored for inclusion in the conservation activities. Although a number of forest conservation studies were conducted in Malawi, very few have addressed the socio-ecological dynamics in remote areas such as the Misuku Hills, where forests are still abundant and small indigenous populations rely heavily on forest resources to meet daily needs.

Future research should also consider examining the dynamics between the Misuku Hills forested region and the highly populated area immediately to the east in northern Karonga. The lake plains to the east are almost completely deforested to support rice cultivation, but it is suspected that the communities in the lake plains rely heavily on the ecosystem services provided by the Misuku Hills. This dichotomy likely creates both opportunities and pressures on the Misuku Hills forested region that generate tensions that need further exploration.

Lastly, a recommendation regarding informed consent. As noted in Section 2.3.1. Ecosystem Services Assessment Photovoice Exercise, the residents of Chikutu were never previously exposed to the informed consent process. This study proved to be an opportunity to teach residents about their rights as subjects of human research and to demand informed consent from future researchers. Any future research should be prepared to confront a lack of awareness of informed consent protocols and empower residents through education about their rights.

#### **5. Conclusions**

Since independence, Malawi has enacted a plethora of forestry policies that have been developed to meet international standards, including policies that promote local control and community-based forest management. What appears to be lacking is the widespread implementation of these policies and the empowerment of communities to practice local control. Despite these challenges, the forest conservation activities in the Misuku Hills demonstrate the potential for rural communities to organize and assume stewardship of forests both on customary land and in public forest reserves. In fact, a focus on communities such as those in the Misuku Hills is necessary to protect the last remaining indigenous forests and the indigenous communities that depend on forests for their livelihood. The ecosystem services inventory conducted in Chikutu revealed that community members have extensive knowledge of forest resources, and these resources are still readily available in communities residing in Malawi's remaining forests.

Malawi continues to be an anomaly on the macro-level economic development pathway. Its poor and rapidly growing population and widespread practice of subsistence agriculture will continue to place pressure on Malawi's forests. Until there is a change in these social forces, an increase in the diversity of livelihood opportunities off the land, and improvements in state institutional capacity, local control and stewardship will be necessary to conserve and regenerate the forests of Malawi.

**Author Contributions:** Conceptualization, C.C.; Methodology, C.C.; Investigation, C.C.; Writing—original draft preparation, C.C.; Writing—review and editing, C.C., T.H., and A.J.

**Funding:** This research was made possible by the US Fulbright Scholar program.

**Acknowledgments:** The authors would like to acknowledge the support of Hassan Milazi of the Ministry of Health of Malawi. This ecosystem services field research would not have been possible without his consultation on the methodological approach and assistance in logistical planning. The authors would also like to acknowledge the hospitality of the Milazi family and residents of Miyombo village that graciously hosted the author during the fieldwork. Information provided too by NGOs in Malawi, Action for Environmental Sustainability and Sustainable Development Initiative, was invaluable to form a complete accounting of conservation activities in the Misuku Hills. Lastly, the faculty of the Department of Built Environment at Mzuzu University were extremely gracious hosts. They provided invaluable guidance on this project and were extremely accommodating in allowing time away from campus to travel to the study site.

**Conflicts of Interest:** The authors declare no conflicts of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript; or in the decision to publish the results.

#### **Appendix A**

#### *Forest Cover Data Discrepancies and Implications*

There have been a number of noted discrepancies in estimates of Malawi's forest cover, rates of deforestation, and forest land tenure designations derived from outdated data sources [6,29]. The presentation of these discrepancies below is meant to instigate a critical evaluation of published forest cover estimates and to shed light on possible recent gains in forest cover (possibly due to enacted forest policies) and current efforts to remedy widely disparate forest cover estimates. It is certainly not meant to close the case on what led to these discrepancies, nor could it as much of the information on estimate sources and methodologies cannot be obtained.

Alarmist statements from as recent as 2019 that "over the last 40 years, more than half of Malawi's forests and woodlands have vanished" [62] (p. 1) are certainly dramatic, but they mask apparent recent gains. Admittedly, these gains are not easily discernable without a more deconstructed examination of trends in Malawi's forest loss over time that reveal a guarded optimism towards recent achievements in reforestation.

One cited estimate from 2016 is that Malawi is losing forest cover at a rate of 2.8% (250,000 ha) annually [27]. The 2.8% assessment appears to have been derived from a 1993 Biomass Assessment Report (assessment conducted in 1991), but this estimate masks significant changes over time and more recent declines in the rates of forest loss. This estimate appears to have been first cited in 2006 by a Forest Conservation Officer at the Food and Agriculture Organization (FAO) Regional Office for Africa. Despite its validity being called into question in 2010 due to its use of data demonstrating a loss of forest cover at a suspiciously unwavering 165,000 ha every 5-year period from 1990–2010 [29], it was still being perpetuated in 2016.

An oft-cited comparison of maps of Malawi forest cover from 1979–1999 reveals a dramatic loss of forest cover over this period, but if any validity can be given to these maps, it appears that this dramatic change happened prior to 1990. Malawi lost 41% of forest cover between 1972 and 1990 at a 2.3% annual rate [6], and this loss of forest cover occurred almost exclusively on customary and private land [97] (as cited in [6]). Post-1990 declines in deforestation rates are being masked when including pre-1990 forest cover loss estimates. Applying the FAO forest classification system and Malawi Forestry Department data (found in Table 8.3 from the cited report), the average annual deforestation rate from

1990–2010 was <1% per year [29]. Another report produced in the same year claims that Malawi lost 0.85% per year or a total of 16.9% (659,000 ha) of its forest cover over this period [98]. The report does not cite the source of this estimate. It may be from the report above, and it is still less than the 1990–2010 1.6% estimate of the same period cited in the introduction of this paper. It appears that pre-1990 estimates of rapid deforestation using a different classification as FAO are being incorporated into more recent estimates and masking 1990–2010 decreases in deforestation rates.

It could be that the rate of deforestation in Malawi after 1990 declined simply because Malawi forests had already been diminished, and there was little forest left to lose. This would be most apparent in the central and south regions, where there is less forest cover. In Dedza district (central region), 1991–2015, almost half of the forest was lost, but it only had 2.6% forest cover in 1991. This was almost a 2% annual loss over this period, but there was not much forest area to lose [15]. This is also pertinent to the districts in the northern region where the Misuku Hills are located. Since independence, there has been a gradual migration of people to the northern region [37], and "it is these areas that have also experienced the greatest amount of deforestation since independence" [14] (p. 274). An 1972–2009 estimate found that Karonga (−579 km2; <sup>−</sup>28%) and Chitipa (−565 km2; <sup>−</sup>20%) were among the districts that experienced some of the greatest declines in forest area [14]. Again, this estimate included pre-1990 conditions that could be masking more recent reductions in forest cover loss. A hot spot of change analysis revealed that, while some districts in the northern region experienced significant forest and natural vegetation loss between 1990–2010, Chitipa and Karonga experienced almost imperceptible changes [5]. These are the data cited in the introduction of this paper as being the most methodologically rigorous and forthcoming.

The 1972–2009 estimate reveals that nationally "there was a loss of 12,760 km2 (36%) of original forested area but also 11,161 km2 of new forest establishment, resulting in a relatively modest overall net loss of 1599 km<sup>2</sup> (5%)" [14] (p. 269). The districts of Chitipa (+54%, +622 km2) and Karonga (+44%, +523 km2) were among the districts that experienced the largest percentage and net gains in mosaic land cover (defined as a mixture of cropland, forest, woodland, grassland, scrubland, and other natural vegetation). In fact, in every district that had an overall loss in forest, there was an overall gain in mosaic land cover. Of course, this calls into question the ability of the mosaic land cover to replace the losses to biodiversity and ecosystem services brought about by the loss of indigenous forest cover.

Other estimates of national forest cover reveal apparent swings. The Biomass Assessment of 1991 showed that, in 1973, "Brachystegia forests occupied 45% of total land area of Malawi (36.5% if Lake Malawi is included) while in 1990/91 land under forest cover was estimated to be 25.3% (20.5% if Lake Malawi is included)" [29] (p. 143). The trusted estimate cited in the introduction of this paper of 26.8% is within the range of other estimates (18.2 to 28.7%) of total forest land cover [99], and greatly underestimates another as high as 34% (3.2 million hectares) [29]. There has undoubtedly been a loss of forest cover since 1973, but there has also been a gain (or at least a stagnation) since 1990. The difference between the 45% estimate in 1973 and the 26.8% estimate in 2010 equates to a 40% loss, but this loss occurred completely before 1990. As cited above, Malawi lost 41% of forest cover between 1972 and 1990 [6]. This is highly contradictory to the 5% net loss cited above that takes into account regeneration and new forest establishment [14]. One explanation that could attempt to reconcile these estimates is that previous estimates did not take into account regeneration, but this is far from clear.

Despite all of the data discrepancies and the opaqueness of how they were derived, the most recent estimates are cause for measured hope. Since the decline in forest cover and deforestation rates coincides with the beginning of the slew of forest policies in the 1990s, the larger unanswered question is the degree to which forest policy interventions led to the apparent halting of deforestation.

The academic exercise and arguable futility in trying to untangle the web of past forest cover and loss estimates are currently being addressed in future plans.

As part of the process for developing Malawi's National Monitoring Framework, the US Geological Survey (USGS) with support from USAID is developing national maps of land use and land cover, as well as maps documenting on-farm tree cover for baseline year 2017, the year the National FLR

Strategy was launched. These maps will provide data on the biophysical progress of FLR interventions in Malawi (e.g., percent of tree cover), which will serve to set a baseline for monitoring biophysical progress on the agricultural technologies, forest management, and community forest and woodlot restoration interventions [100] (p. 16)

These data are not just essential for forest monitoring, but also because they will inform other monitoring activities such as the Integrated Household Survey which considers community forests and woodlots in its assessment of local resources [100].

"In 2016, the Government of Malawi made a national pledge to the African Forest Landscape Restoration Initiative (AFR100) under the Bonn Challenge to restore 4.5 million hectares of degraded and deforested land by 2030" [100] (p. 3). Malawi has taken the bold step to not only make this pledge and have an increasing number of policies to support this goal, but Malawi now also has a 2018 framework to evaluate its progress towards reaching its pledge to "increase area of community forests and woodlots to 200,000 ha by 2020 and 600,000 ha by 2030" and "improve protection and management of 2 million ha of natural forest, restore 500,000 ha of degraded forest, and establish 100,000 ha of commercial plantations by 2030" [100] (p. 4).

#### **References**


© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

### *Article* **Deforestation and Connectivity among Protected Areas of Tanzania**

#### **Belachew Gizachew, Jonathan Rizzi, Deo D. Shirima and Eliakimu Zahabu**


Received: 30 December 2019; Accepted: 27 January 2020; Published: 4 February 2020

**Abstract:** Protected Areas (PAs) in Tanzania had been established originally for the goal of habitat, landscape and biodiversity conservation. However, human activities such as agricultural expansion and wood harvesting pose challenges to the conservation objectives. We monitored a decade of deforestation within 708 PAs and their unprotected buffer areas, analyzed deforestation by PA management regimes, and assessed connectivity among PAs. Data came from a Landsat based wall-to-wall forest to non-forest change map for the period 2002–2013, developed for the definition of Tanzania's National Forest Reference Emissions Level (FREL). Deforestation data were extracted in a series of concentric bands that allow pairwise comparison and correlation analysis between the inside of PAs and the external buffer areas. Half of the PAs exhibit either no deforestation or significantly less deforestation than the unprotected buffer areas. A small proportion (10%; *n* = 71) are responsible for more than 90% of the total deforestation; but these few PAs represent more than 75% of the total area under protection. While about half of the PAs are connected to one or more other PAs, the remaining half, most of which are Forest Reserves, are isolated. Furthermore, deforestation inside isolated PAs is significantly correlated with deforestation in the unprotected buffer areas, suggesting pressure from land use outside PAs. Management regimes varied in reducing deforestation inside PA territories, but differences in protection status within a management regime are also large. Deforestation as percentages of land area and forested areas of PAs was largest for Forest Reserves and Game Controlled areas, while most National Parks, Nature Reserves and Forest Plantations generally retained large proportions of their forest cover. Areas of immediate management concern include the few PAs with a disproportionately large contribution to the total deforestation, and the sizeable number of PAs being isolated. Future protection should account for landscapes outside protected areas, engage local communities and establish new PAs or corridors such as village-managed forest areas.

**Keywords:** deforestation; isolation; protected areas; buffer areas; Tanzania

#### **1. Introduction**

Overwhelming evidence shows that Protected Areas (PAs) in the form of either National Parks, Forest Reserves or other forms of protection have lower deforestation rates than unprotected landscapes [1–5]. Establishing and managing protected areas is, thus, one of the most important policy tools for achieving environmental, natural resource conservation and climate goals. Following this premise, the United Nations Convention on Biological Diversity [6] recommends each country establish and manage protected areas to conserve biological diversity. The United Nations 2030 Agenda for Sustainable Development and its Sustainable Development Goals (SDGs) recognize protected areas as a key strategy for biodiversity conservation and sustainable development in the targets

they contain, such as the Aichi Biodiversity Target 11, SDG goals 14 and 15 [7]. Furthermore, an opportunity was presented for conservation of tropical forests through the United Nations Climate Agreement [8] on reducing emissions from deforestation and degradation, plus forest management and conservation, and enhancement of carbon stock (REDD+). Therefore, the most important immediate steps to achieve these goals include intensive conservation of existing protected areas, establishing additional conservation areas of tropical forests, and supporting areas of high conservation benefits in terms of carbon, biodiversity and other ecosystem services.

The United Nations [9] records more than 200,000 protected areas worldwide, of which 840 are in Tanzania. Tanzania indeed devoted a sizeable proportion of its land area (36%) for conservation, with an original goal of conserving forests, landscapes and wildlife. PAs in Tanzania are currently managed most commonly by the central government or local authorities, either as Forest Reserves, Game Controlled areas, Game Reserves, National Parks, Nature Reserves, Village Forest Reserves or Forest Plantations. Tanzanian PAs include some UNESCO world heritage sites such as the Kilimanjaro and Serengeti National Parks, Selous Game Reserve and Ngorongoro Conservation Area; and series of PAs with exceptional endemism along the Eastern Arc Mountains. Tanzanian PAs are generally regarded as biodiversity hotspot, with over 10,000 plant species, hundreds of which are nationally endemic. Of the plant and animal species, 724 are identified as "threatened" in the Red List of the International Union for Conservation of Nature (IUCN), with 276 species classified as "endangered" [10].

Other than biodiversity conservation, PAs in Tanzania are offering an increasingly diverse set of ecosystem services. Among them is the significant contribution to the national economy through tourism revenues, most popular of which is ecotourism, involving natural environments and wildlife, through which Tanzania remained the best safari destination in Africa. As a result, Tanzania's tourism sector is one of the most significant income earners. Furthermore, the sheer size (total area) and the large number and diversity of PAs in Tanzania means that their role in mitigation to climate change through carbon sequestration, and thus the potential to garner financial benefits, is enormous.

Deforestation caused by human activities such as agricultural expansion, charcoal production and illegal logging inside and within the buffer areas can undermine the ecosystem and climate benefits of PAs. Deforestation in buffer areas further undermines the connectivity among PAs, and thus lead to isolation [11], which in turn can potentially cause restriction of the ability of plant and animal species to relocate to new geographic areas as well as changing plant community structure and diversity within PAs because of herbivore concentration [12]. Consequently, conservation and connectivity of PAs have international significance as the Aichi Target 11 of the Convention on Biological Diversity demands countries have at least 17% of the land covered by well-connected PA systems by 2020 (IUCN). The Millennium Ecosystem Assessment [13] has long identified deforestation as the primary driver of biodiversity loss. Therefore, reducing deforestation and improving the connectivity of PAs play fundamental role to ensure species survival, particularly in the context of habitat protection. Habitat fragmentation and isolation that can be caused by anthropogenic activities obstruct the possibility for genes and species to move amongst protected areas [14].

Annual deforestation in Tanzania was close to 470,000 ha between 2002 and 2013 [15], which constitutes a significant contribution of the total anthropogenic emissions from the land-use change in the country. These would provide the theoretical basis for strengthening the protection of existing conservation areas, allocating additional areas of conservation and improving the connectivity of PAs. Tanzania acknowledges deforestation as a major threat to biodiversity and ecosystem services, and is committed to most of the targets in the Convention on Biological Diversity [16]. This includes a commitment to effectively manage existing protected areas (Target 11), and to significantly reduce the rate of degradation and fragmentation of ecosystems and the loss of habitats (Target 5). However, forest cover loss due to deforestation inside PAs and in their external buffer areas are often not objectively quantified and analyzed, particularly given the vast size and number of PAs in Tanzania. Studies that quantify deforestation inside of PAs and assess connectivity and isolation would be useful to understand the climate change mitigation potentials and the conservation benefits of PAs.

This study draws on the best available Landsat based remote sensing data of forest cover change for the period 2002–2013, to assess deforestation within PAs and their external buffer areas in Tanzania. The specific objectives are (1) to evaluate deforestation inside and within the buffer areas of PAs, (2) evaluate connectivity between PAs, the lack of which can potentially lead to isolation, and (3) whether and to what extent deforestation among PAs vary by PA management regimes.

#### **2. Data and Method**

#### *2.1. Study Area*

The study area covers the mostly centrally managed PAs and those located in mainland Tanzania where they are collectively known as Conservation Areas. These are sub-grouped based on management regimes as National Parks (NP), Game Reserves (GR), Game Controlled (GC) areas, Nature Reserves (NR), Forest Reserves (FR), and Forest Plantations (FP) (Figure 1). General characteristics of the management regimes are summarized in Table 1. We excluded PAs designated as village forest reserves and wildlife management areas, due to inadequate spatial coverage data and lack of accurate shape polygons. In addition, we also removed 48 small PAs (area = 0–10 ha) located either on islands or mostly mangroves near the cost as they were not adequately covered by either the shape files or the deforestation map. Additional areas that are excluded from the analysis include buffers overlapping with water bodies or territories of other countries. The study finally included 708 PAs covering a total area of nearly 31 million ha and their corresponding unprotected buffer areas of more than 60 million ha within a range of 0–10 km surrounding each PA.


**Table 1.** Characteristics of the six Protected Area (PA) management regimes in main land Tanzania.

#### *2.2. Data*

PA polygons: we obtained data for the location and boundary polygons of the 708 PAs from the World Database for Protected Areas (WDPA) updated January 2019 [17], managed by the United Nations Environment World Conservation Monitoring Centre (UNEP-WCMC) with support from IUCN and its World Commission on Protected Areas (WCPA) (Protectedplanet.net).

Deforestation data: we extracted data for 11 years (2002–2013) of deforestation area for all PAs and their buffer areas from a wall-to-wall deforestation map of Tanzania, developed for the purpose of Forest Reference Emission Level (FREL) of Tanzania (FREL) [15]. The wall-to-wall deforestation map was developed from changes from Landsat 7 Enhanced Thematic Mapper Plus (ETM+) (2002) to Landsat 8 Operational Land Imager (OLI) (2013), covering the entire mainland Tanzania. The deforestation map used a total of 85 Landsat 7 ETM+ (2002) and Landsat 8 (2013) images, with a resolution of 30 m. The Landsat 7 ETM+ scenes were pre-processed by the USGS to surface reflectance level using the Landsat Ecosystem Disturbance Adaptive Processing System (LEDAPS) atmospheric and topographic correction algorithm. The Landsat 8 scenes were pre-processed to surface reflectance level by the United States Geological Survey USGS internal L8SR algorithm.

**Figure 1.** Location of Tanzania in Africa and Protected Areas (PAs) in Tanzania by management regimes. The different colors (except the blue which are water bodies) represent management regimes, namely (in no particular order), National Parks (NP), Game Reserves (GR), Game Controlled (GC) area, Nature Reserves (NR), Forest Reserves (FR), and Forest Plantations (FP).

The advantage of the wall-to-wall deforestation data used here over other change maps such as Global Forest Watch include, (1) Forest definition used for classification was based on predetermined national forest definition based on forest area (at least 0.5 ha), crown cover (at least 10%) and potential tree height (at least 3 m); (2) the land-use classification was monitored and evaluated by expertise with knowledge of the area, and (3) the accuracy of the deforestation map was evaluated using a combination of the National Forest Inventory (NAFORMA) plot data of the 2010 and the Regional Centre for Mapping and Resource Development (RCMRD) Land use land cover map of Tanzania.

We extracted deforestation data as land-cover change from forest to non-forest (ha) for all PAs for the period 2002–2013. We also extracted deforestation data for pairs of internal and external buffers in concentric bands of 0–0.5 km, 0.5–1 km, 1–5 km and 5–10 km measured from the boundary of each PA. Internal buffers here after refer to areas just inside the boundaries of the PAs towards the center of the PAs, while external buffers refer to areas just outside the boundaries of the PA. For the construction of the concentric bands in the external buffer zones, areas that fall either on another PA, water bodies, or outside of the territories of Tanzania were excluded. For those PAs where the internal concentric bands were not possible to construct (i.e., size of the PA smaller than the concentric band area), the

deforestation for that concentric band was estimated as the deforestation of the entire PA. Where PAs overlap, the areas are merged for total area estimation. Figure 2 shows the work flow in ArcGis on the simultaneous construction of concentric bands, and extraction of data on each PA and its corresponding concentric bands from the deforestation data, and thweir export to a worksheet. In the absence of data from field observation or measurements of the period, only visual validation was made using independent images from space imagery providers including Google Earth, Environmental Systems Research Institute (ESRI) and DigitalGlobe for selected PAs and buffer areas.

**Figure 2.** Work flow diagram in ArcGis for the simultaneous construction of concentric bands and extraction of data and output to worksheet.

#### *2.3. Statistical Analysis*

*Deforestation inside and within the bu*ff*er areas of PAs*: deforestation was defined as an area converted from forest to non-forest (ha) within each PA and estimated as the sum of the areas of individual pixels with forest-to non-forest conversion during the period 2002–2013. Proportion of the area deforested was then estimated as (a) the ratio of the area deforested to the total area of the PA, which indicates the absolute forest to non-forest conversion rate; and (b) the ratio of the area deforested to the total area of the forest within the PA at the beginning of the monitoring period, which indicates the relative forest to non-forest conversion rate. Further, proportion of the area deforested within each buffer area was calculated as the ratio of area deforested within the given concentric band (0–0.5; 0.5–1; 1–5; 5–10 kms) divided by the area of that concentric band. The metric allows pairwise comparisons and correlations analysis between the rates of deforestation inside the boundaries and that of unprotected buffer areas.

*Connectivity and isolation of PAs*: given the lack of indicators or quantitative criterion to define connectivity or the lack of it (isolation); we identified PAs surrounded by unprotected landscapes, and no connection to the neighboring PA(s) within at least 1 km from their boundaries. We identified such PAs as "isolated" and assessed their unprotected external buffer areas for deforestation in an increasing distance within 0–0.5; 0.5–1; 1–5; 5–10 km from their boundaries. Pairwise *t*-test and correlation analysis were used between deforestations inside boundaries of isolated PAs and the corresponding buffer areas with increasing distances to assess the pressure of activities outside the PA boundaries on the corresponding PAs.

*PA management e*ff*ects*: we used the generalized linear model (GLM) for the analysis of variance (ANOVA) to test the variations in deforestation among the six PA management categories (FP, FR, GC, GR, NP, and NR). Since there is a considerable size variation among the PAs, the GLM considered the area-weighted mean. An alternative approach was an ANOVA accounting for PA management, PA size, and PA management by PA size interaction, but this produced the same results and thus the latter was omitted. Following ANOVA, the Duncan's multiple range test was used to compare the area-weighted mean deforestation among the six management categories. PA size effects entered the

analyses by dividing PAs into three size-based cohorts. After preliminary tests, three percentile-based cohorts were found to be sufficient. These are PAs with size less than or equal to the 25th percentile (Q1); PAs that are larger than the 25 percentile and less than or equal to the 75 percentile (Q2); PAs that are larger than the 75th percentile (Q3). These cohorts were further used to test whether PAs in the smaller cohorts are disproportionately deforested than those in the larger cohorts.

#### **3. Results**

#### *3.1. Deforestation inside Protected Areas (PAs) and the Bu*ff*er Areas*

Annual deforestation averaged 140 422 (STD = 922) ha during 2002–2013, which in absolute terms amounts to 5% of the total land area of all PAs combined. The corresponding annual forest loss within the PAs was about 0.8%. At the individual PA level, deforestation varied widely among PAs. About 23% (*n* = 160) of the PAs received effective and fortress type protection and thus no deforestation. This includes most National Parks and series of Forest Reserves, such as PAs along the Usambara Mountains (Figure 3). In contrast, some PAs have lost more than 50% of their forested areas during the same period (e.g., Makere South Forest Reserve). Deforestation was rather concentrated in few, larger PAs. A small proportion (10%, *n* = 71) of the PAs contributed more than 90% of the total deforestation during the monitoring period. However, in terms of land area, these 71 PAs represent 77% of the total protected areas. Inside deforested PAs, more deforestation in general occurred near the periphery of their boundaries than the interior. Comparing deforestation rates in the inside peripheries of PAs and their external buffer areas, 51% (*n* = 359) of the 708 PAs exhibited significantly lower deforestation rates. Table 2 summarizes the characteristics of deforestation among PA management regimes. Figure 3 visually demonstrates selected PAs representing highly protected and highly deforested PAs; and Figure 4 contrasting protection inside and the buffer areas.


**Table 2.** Summary of PA characteristics: number and size of PAs, total 11 years (2002–2013) deforestation (ha), and area deforested as percentage of PAs total management area, and PA forest area at the beginning of monitoring period.

**Figure 3.** Series of PAs along the Usambara Mountains of northeastern Tanzania along the eastern most ranges of the Eastern Arc Mountains. Note: effective protection with few spots or no deforestation inside and the buffer areas of the PAs. The buffer area in the figure is 1 km non-overlapping zone surrounding the PA network. Panel (**A**) is. of this study; and Panel (**B**) is image from space imagery providers.

**Figure 4.** Example of PAs with negligible deforestation inside their boundaries (Panels (**C**,**D**), Karitu Forest Reserve) and with a sizeable and advancing deforestation inside the boundary (Panels (**A**,**B**); Ugala North Forest Reserves), in western Tanzania.

#### *3.2. Connectivity and Isolation of PAs*

The analysis on connectivity showed that 352 PAs are at least 0.5 km away from their neighboring PAs while 293 PAs are at least 1 km away from the nearest PA. In terms of management regime, these isolated PAs mostly belong to the Forest Reserves, consistent with their higher number among the different regimes. In general, in the buffer areas of PAs, deforestation rates were lower in areas closer to the external boundaries (in the buffer zone 0–1 km) than those further away, e.g., in the buffer zones 5–10 km (e.g., Figures 5 and 6). Deforestation in buffer areas as percent of the unprotected buffer areas in a range of 0–10 km was estimated at 7%. We found a strong and positive correlation (*p* < 0.0001) between deforestation inside the boundaries and deforestation in the buffer areas, although declining with distance. Figure 5 shows two isolated PAs (Matogoro West and Matogoro East), where deforestation inside is highly correlated with deforestation in the outside buffer areas. A series of PAs across Tanzania, for example those located along the central part of the Eastern Arc Mountains, shows evidence of lack of connectivity, which we defined as an indicator of isolation (Figure 6).

**Figure 5.** Deforestation in the buffer areas of two isolated PAs (Forest Reserves of Matogoro West and Matogoro East), Southern Tanzania. Panel (**A**) is of this study; and Panel (**B**) is image from space imagery providers.

**Figure 6.** Deforestation inside and the buffer areas of mostly disconnected PAs (Forest Reserves) along the Eastern Arc mountains in Eastern Tanzania. PAs in figure are (North to South): Nguru North, Mamboto, Mkongo, Nderema, Pumula, Mbwegere, and MKuli Forest Reserves. Panel (**A**) is. of this study; and Panel (**B**) is image from space imagery providers.

#### *3.3. PA Management E*ff*ects*

Results of analysis of variance (ANOVA) revealed that the differences in area-weighted mean deforestation among PA management regimes are significant (*p* < 0.0001). An alternative ANOVA that considered PA management categories, PA size and interaction effects also revealed significant (*p* < 0.0001) management, size and interaction effects. Following ANOVA, the results of Duncan's means grouped the six management categories into three. Game Controlled Areas presented the highest area-weighted mean annual deforestation; followed by Forest Reserves, Game Reserves and National Parks; while Forest Plantations, and Nature Reserves exhibit the lowest (Figure 7). Therefore, most of the deforestation in Tanzanian PAs during the period occurred in Game Controlled Areas and Forest Reserves. However, this does not mean that deforestation is high in all Game Reserves or Forest Reserves; given the large variations within protection status of same management regime (Figure 7). The percentage of deforestation during the entire 11 years (deforestation as percent of total PA area) is the highest in Forest Reserves, followed by the Game Controlled areas while the lowest was in Nature

Reserves. Controlling management effects, deforestation was significantly small for the lowest size cohort (Q1) (*p* < 0.05), while there is no significant difference between the median (Q2) and the higher cohort (Q3).

**Figure 7.** Area-weighted annual deforestation rates (ha/yr) in protected areas of Tanzania (*n*/; = 708) and the six PA management regimes namely Forest Plantations (FP), Forest Reserves (FR), Game Controlled areas (GC), Game Reserves (GR), National Parks (NP) and Nature Forest Reserves (NR).The central notched line is the median, and the diamonds are the area-weighted mean; and the whiskers are the lower and the upper confidence limit.

#### **4. Discussions**

#### *4.1. Deforestation and Isolation of PAs*

The estimated annual deforestation rate of 0.45% among PAs during the 11 years of the monitoring period is a significant reduction compared to an estimated annual deforestation rate of 0.63% for the unprotected buffer areas. This is further strengthened by the finding that two-thirds of the PAs have significantly less deforestation than the surrounding landscapes, including those with no or negligible deforestation inside their boundaries. This includes some of the famous National Parks and Nature Reserves with little or no deforestation, being inaccessible and fortressed. This suggests PAs have contributed in reducing the otherwise ferocious annual deforestation rate of 0.7% in Tanzania during the same period, 2002–2013 [15]. Therefore, protection can be an effective strategy for reducing deforestation, consistent with some recent studies [18]. These relatively well protected PAs represent 22% of the land area of Tanzania; meaning Tanzania has already succeeded in achieving the 17% target of its territory as protected defined by the Convention on Biological Diversity [16]. This demonstrates Tanzania's commitment to conservation as a signatory of major international biodiversity treaties, and by implementing domestic conservation laws and regulations enshrined in the National Forest Act of 2002.

However, our results showing some PAs that are deforested as much as the unprotected buffer areas and some that have lost up to 50% of their forested area during the 11 years indicates that not all PAs received similar protection status. This is common in a number of rainforest areas and regions with

large number of protected areas, where protection resulted in mixed outcomes in terms of reducing deforestation [2,3,5,19–22]. Uniquely for Tanzania, however, only few large PAs (*n* = 75) represent the great majority (>90%) of deforestation. These highly deforested PAs should remain a management concern, because despite their number, they represent more than two thirds of the protected area in Tanzania in terms of size.

On the other hand, more than 50% of Tanzanian PAs are well connected to one or more neighboring PAs, particularly those in western Tanzania connected to Katavi national park and those in the south western Tanzania surrounding the Selous Game reserve (Figure 1). This makes Tanzania one of the few countries fulfilling the Aichi Target of "well-connected" PAs [16]. While these results may be encouraging, we also found the remaining half, most of which are Forest Reserves (371 Forest Reserves), are isolated by at least 0.5 km, often surrounded by land use or landcover other than forests. Other studies [18,23] showed that unprotected landscapes adjacent to many protected areas have been converted to other land uses. In many cases, isolation becomes the reason for deforestation to push into the boundaries of PAs, threatening the effectiveness of PAs to maintain viable forest and protect biodiversity in the long term [14].

The significantly strong correlation between deforestation in the buffer areas and deforestation inside PAs shown in this study suggests that PAs are being influenced by human activities outside their boundaries. In particular, Forest Reserves and Game Controlled areas that are located near or inside of human dominated landscapes are subject to isolation. For instance, PAs along the Eastern Arc mountains are known for exceptional biological and conservation importance but have long been threatened by deforestation [24,25]. Our observation corroborates these previous reports, albeit the different monitoring periods, that large parts of the buffer areas of many PAs along the Eastern Arc Mountains were deforested (e.g., see Figures 5 and 6) or isolated, although well protected (Figure 3) which, in the absence of intervention, can potentially lead to further isolation and encroachment into PA territories, threatening the ecosystem services they can provide. Nevertheless, the potential for connecting those isolated PAs is immense, because most are located within short distances from each other. For instance, along series of PAs in the Usambara Mountains of northeastern Tanzania (Figure 3), developing a 1 km corridor surrounding each PA can effectively connect nearly all protected areas in that mountain range. Indeed, as shown in Figure 3, effective protection inside PAs might have prompted protection across the unprotected buffer zones surrounding those PAs. Remaining governance challenges could be designing and implementing compensation schemes for conservation-related displacements of people in buffer areas and sometimes inside PA territories [26].

#### *4.2. PA Management E*ff*ects*

Knowledge on the degree to which responsible public institutions can protect their respective natural forests and biodiversity will have a profound importance to the public and Tanzanian decision makers. At the level of PA management, Forest plantations, National Parks and Nature Reserves exhibit significantly less deforestation rate than Forest Reserves and Game Controlled areas (see also Figure 7). These results are strikingly similar to that of Uganda [22], in which Forest Reserves lost forest carbon while National Parks and wildlife Reserves gained forest carbon during 10 years monitoring period. Another study in East Africa [27] suggested that the other management regimes performed poorly as compared to National Parks in reducing deforestation. National Parks and Nature Reserves may have better protection status than Forest Reserves, most likely because they benefit from inaccessibility and fortress-type protection assisted by tourism revenues to support and strengthen protection. While these are generalized conclusions, we also see that there is a large variation in protection status within the same management regime (Figure 7). For instance, while Forest Reserves in general were deforested, there are a series of Forest Reserves, for instance in Eastern Usambara Mountains in northeastern Tanzania that are well protected and showed no or little deforestation during the period. This result is of significance because the Usambara mountains make up the Eastern Arc forests which have the highest known number of plant and animal species of any region in Tanzania.

The observed deforestation, particularly in the top 10% highly deforested PAs dominated by Forest Reserves is most likely attributed to their location leading to an increasing external pressure associated with the increasing human population in the surrounding landscapes. Giliba et al., [28] suggested that the threat against Forest Reserves is well connected to an increasing demand for household energy and the need for new land for cultivation and settlements near population areas. Visual observations show that even when PAs are well protected, they may be surrounded by recently deforested landscapes (e.g., see Figures 4–6). Such fortress-type protection may cause leakage (spillover) to neighboring buffer areas [29], and that leakage might accelerate the rate at which PAs become isolated [23]. Therefore, PA management need to consider the potential of leakage. However, given the limited field observation, and lack of histories of the PAs on their establishment, we could not ascertain whether leakage was responsible for the observed isolations in Tanzanian PAs.

PAs may be established purposely in dense forests, higher elevations, steeper slopes or long distances to roads and settlements, particularly those that have been established many years ago. Our comparison between changes inside the boundaries and the buffer areas did not consider the possible differences in land characteristics and possible biases of locations during the establishment of these PAs. However, with a current expanding infrastructure and fast-growing young population demanding agricultural land, most of the PAs are within the reach of human activities. Some of the urgent measures for management authorities and other stakeholders should therefore include reviewing existing management approaches, to consider participatory management which promotes partnerships and offers benefit sharing and other development opportunities to communities living outside PAs. Successful practices of engaging local communities exist in Tanzania and experiences can be drawn from the past participatory forest resource-management programs [30], and carefully adapted to serve PA management objectives.

#### *4.3. Implications for the Climate Benefits of Protection*

Protection in Tanzania has historically been intended for ecosystem services, such as ecotourism and biodiversity conservation. More recently, the Nationally Determined Contributions (NDCs) under the United Nations Framework Convention on Climate Change (UNFCCC) [8] entails low greenhouse gas emissions and climate-resilient development. In particular, policies such as REDD+ recognized conservation as one of the five major activities [8], providing additional opportunity for PAs. More specifically, Tanzania's recent Climate Smart Agriculture (CSA) guideline [31] recognizes landscape and ecosystem services and payments for ecosystem services as key to achieving its sustainable development goals. Furthermore, as part of its commitment under African Forest Landscape Restoration (AFR100) initiatives, in 2018 Tanzania pledged to restore 5.2 million hectares of degraded and deforested land by 2030. Tanzania can thus use this opportunity to select those buffer zones or corridors as restoration areas under such programs.

Given the large sizes and diversity of PAs, protection in Tanzania can make significant and vital contributions to emissions reduction. This study also provided evidence that strictly protected PAs are effective at reducing forest losses and thus reducing emissions. However, protection is particularly challenging and resource-intensive in countries such as Tanzania, with high forest dependence where forest-based charcoal and fuel wood are the single most important sources of household energy [32], and forest lands provide the last remaining lands for agricultural expansion [33].

Initiatives such as REDD+ and other national forest-management strategies are expected to provide a solution through providing incentives for the respective authorities and to the local communities. Consequently, Tanzania can benefit from PA managements as a national strategy and policy options to achieve its climate goals, through reducing emissions from deforestation and forest degradation inside the PAs. Protection should also consider the buffer areas through, for instance, initiating actions to restore and develop unprotected areas into corridors, through initiatives such as promoting community forest reserves, landscape restoration and conservation agriculture, and improving the connectivity of isolated PAs. Improving connectivity requires strong cooperation and partnership between the

different PA management regimes, developing approaches on how communities living adjacent to PAs can participate and share the benefits. Community engagement and benefit sharing can avoid the pitfalls of the current management in which several Forest Reserves appear isolated, surrounded by deforested or landscapes.

#### **5. Conclusions**

This study provided a quantitative assessment of deforestation in all the major PAs and their corresponding buffer areas across six PA management regimes in Tanzania. Such knowledge will contribute to understanding of the conservation and the climate change mitigation potentials of PAs. We see that the outcomes of protection in Tanzania are generally mixed, ranging from fortress-type protection with no deforestation detected to those PAs where protection did not significantly reduce deforestation. Yet, most PAs in Tanzania have been effective in reducing deforestation, despite the significant land-use pressure from outside their territories. This provides considerable support to the notion that protection remains one of the most effective policy tools to reduce deforestation, and thus protecting valuable landscapes and biodiversity, conforming to the original goal of protection. These results also demonstrate the potential that PAs can offer a considerable opportunity to achieve long-term climate goals such as the nationally determined contributions (NDCs) and climate mechanisms such as REDD+.

Despite successes in some PAs across the different management regimes, there is a clear need to strengthen protection of the few but large-area PAs with high deforestation rates and promoting connectivity of many isolated PAs through forest landscape restoration and developing corridors. The challenges are often designing and implementing win-win solutions both for the PA management and communities that can be affected by conservation. Successful participatory forest management programs that engage communities exist in Tanzania, from which experiences can be drawn and adapted to PA management.

We demonstrated the utility of a wall-to-wall land-cover change map to monitor deforestation inside PAs and the unprotected buffer areas at various distances. Given the dynamics of deforestation in Tanzania, however, there is a need for updated data and supplementary field observation to improve the utility of such data and assessments for a more assertive recommendation.

**Author Contributions:** B.G.: Study design, Methodology, Formal Analysis; Investigation, Original draft Preparation, Writing, Review & Editing. J.R.: Data mining and analysis, methodology, Writing Review and Editing; D.D.S.: study design, validation, writing, review and Editing; E.Z.: Study design, Review and Validation. All authors have read and agreed to the published version of the manuscript.

**Funding:** The Norwegian Institute of Bioeconomy Research and the National Carbon Monitoring Center of Tanzania supported the study.

**Acknowledgments:** We acknowledge support from the Norwegian Institute of Bioeconomy and National Carbon Monitoring Center of Tanzania. Reviews of two anonymous reviewers improved the article.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

### *Article* **Variation in Deadwood Microsites in Areas Designated under the Habitats Directive (Natura 2000)**

#### **Leszek Bujoczek, Stanisław Zi ˛eba and Małgorzata Bujoczek**


Received: 28 March 2020; Accepted: 23 April 2020; Published: 25 April 2020

**Abstract:** The continuing decline in biodiversity presents a major environmental protection challenge. The conservation of sufficiently extensive and diverse habitats requires an array of coordinated actions, often involving large areas. While a set of conservation objectives have been defined for the Natura 2000 network, no universal methods of accomplishing them have been specified, and so they must be designed by individual Member States. Deadwood volume and the density of large deadwood pieces are widely used for evaluating the quality of forest habitat types designated under the Habitats Directive. In the present study, data from 5557 sample plots were used to evaluate the mean values of the two deadwood indicators as well as the ratio of deadwood volume to living tree volume for each of the 13 habitat types in Poland. In addition, a logistic regression model was constructed to evaluate the effects of terrain, site, and tree stand characteristics as well as protection type on deadwood volume in Natura 2000 areas. Mean deadwood volume varied greatly between habitat types, with the lowest values found for Central European lichen Scots pine forests (91T0–2.5 m3 ha<sup>−</sup>1) and Old acidophilous oak woods (9190–4.4 m<sup>3</sup> ha<sup>−</sup>1), and the highest for Riparian mixed forests (91F0–43.1 m3 ha<sup>−</sup>1) and Acidophilous *Picea* forests of the montane to alpine levels (9410–55.4 m<sup>3</sup> ha<sup>−</sup>1). The ratio of deadwood volume to living tree volume ranged from approx. 1%–17%. Additionally, the presence of large deadwood differed among habitat types: in some, there were no deadwood pieces with a diameter of <sup>≥</sup>50 cm, while their maximum density was 6.1 pieces ha<sup>−</sup>1. The logistic regression model showed that the likelihood of a habitat type to have a 'favorable conservation status' as defined by deadwood abundance (a threshold of at least 20 m<sup>3</sup> ha−<sup>1</sup> according to Polish manuals on habitat type evaluation) increased with sample plot elevation, site fertility, and moisture, as well as stand age and volume. Positive effects were also observed for forests under strict and active protection versus managed forests. Planned efforts are necessary to enhance the quality of habitats with insufficient deadwood, especially in managed forests. Special attention should be given to areas that are readily accessible due to gentle terrain and low site moisture. Furthermore, younger stands on less fertile sites may require intervention to promote deadwood accumulation. We recommend retaining a certain proportion of mature stands until natural death and decomposition. Increasing the density of large deadwood is currently one of the most pressing conservation needs in most habitat types.

**Keywords:** reserve network; biodiversity; large trees; snags; coarse woody debris; regression model; habitat conditions; strict protection; managed forests

#### **1. Introduction**

Europe boasts the largest network of coordinated conservation areas in the world, known as Natura 2000, which covers more than 18% of the land area of EU Member States and almost 10% of their territorial waters. Its overarching objective is to ensure long-term conservation of valuable and endangered species and habitats [1–5] in line with the EU legislation including the 1979 Birds Directive [6] and the 1992 Habitats Directive [7]. As far as the habitat types defined under the latter directive are concerned, this means that their range should be preserved (and possibly expanded) and that they should retain their specific structure, functions, as well as characteristic species. The Natura 2000 network extends protection to a total of 231 habitat types; the greatest number of which are forest habitats (81), accounting for half of the entire Natura 2000 area [8]. The EU's special interest in forests is attributable to the fact that natural and semi-natural woodlands are among the most biologically diverse ecosystems on Earth. Although more than 1/3 of Europe is covered with forests, only 10% of their area consists of natural or semi-natural stands that significantly contribute to preserving plant and animal species [4]. While the Habitats Directive [7] laid out a number of conservation goals, it did not indicate specific ways of achieving them, with each Member States being responsible for developing its own solutions [9].

The Habitats Directive [7] requires the countries participating in the Natura 2000 network to monitor the conservation status of the natural habitats and species listed in its appendixes. The monitoring of forest sites encompasses a number of elements, such as their range, area, the species composition of all forest layers, species provenance, the age and vertical structure of vegetation, as well as soil and water conditions. It is essential to identify the factors and threats affecting a given habitat type, such as those associated with forest management [10], which could deteriorate habitat quality and decrease biodiversity. In this context, of importance is a sufficient presence of diverse microsites offered by so-called biocenotic trees, large trees, and dead trees [11,12]. Importantly, as many as 25% of species found in forests are facultatively or obligately associated with deadwood, and some of them are among the most endangered organisms of European temperate forest ecosystems [13]. Their presence is dependent not only on the quantity, but also the quality of deadwood, such as species, diameter, and decay stage [14,15]. Therefore, the removal of dead and dying trees is perceived as detrimental to most forest habitats and is being monitored on Natura 2000 sites [10]. The adoption of an appropriate deadwood management strategy requires knowledge about the ecology of saproxylic organisms, including the size and dispersal of their populations. Such information is needed to decide whether deadwood volume should be increased evenly across the entire managed forest area, but only to a limited extent, or perhaps the focus should be on a substantial improvement in the number and diversity of deadwood microsites in selected areas [16]. Investigations aiming to determine deadwood thresholds have indicated a wide range of desirable deadwood volumes [17]. Threshold values for different habitat types are also provided in guidelines for assessing Natura 2000 sites [18,19].

Currently, in Poland there are 849 special areas of conservation (SACs) with an overall area of 3.9 million ha. They represent 77 habitat types listed in Annex I to the Habitats Directive. Seventeen of them are priority habitat types, whose range is mostly or exclusively limited to the territory of the EU Member States, and so their survival depends directly on the conservation efforts undertaken by those countries. The natural habitat types identified under the Habitats Directive are classified into aquatic and waterside, heath and scrub, meadow, grassland, boggy, rocky, as well as typical forest categories [20]. This work focuses on forest habitats, and in particular on the deadwood they contain as a factor strongly affecting biodiversity. The study involved data derived from several thousand sample plots located on 312 Natura 2000 sites distributed throughout Polish lowlands, uplands, and mountains. Deadwood volume and the density of large deadwood pieces were analyzed with respect to guidelines specifying their threshold values in habitats. Logistic regression was used to determine which site, stand, and protection parameters had a significant effect on deadwood accumulation in those habitats. The results could be helpful in designing appropriate actions and strategies to improve habitat quality.

#### **2. Materials and Methods**

#### *2.1. Data Collection*

As part of the National Forest Inventory (NFI), Poland is covered with a 4 × 4 km grid of sample plots based on the 16 × 16 km ICP Forests network used in the European Union to evaluate forest damage [21]. At each node, there is an L-shaped cluster of five sample plots spaced 200 m apart (Figure 1). The exact rules for taking measurements and general reports on forest conditions are presented in ME [22] and NFI [23]. Depending on the age of the dominant tree species, the sample plots range in size from 200 to 500 m2. In the present study, diameter at breast height (DBH) was measured for standing living and dead trees, while in the case of snags (with a height of at least 0.4 m) and downed deadwood, diameter was measured halfway along their height or length, respectively. Only standing trees with a DBH of >7 cm and downed deadwood fragments with a diameter of >10 cm at the thicker end were included in the study. Measurements were conducted for those dead trees or their fragments that grew within the sample plots prior to death. Thus, the entire length of deadwood fragments lying across sample plot borders was included. On the other hand, stumps left from management procedures were excluded.

**Figure 1.** Sites designated under the Habitats Directive and the layout of a cluster of sample plots.

Forest ecosystems were evaluated on the basis of data obtained in the years 2010–2014 from NFI sample plots located within Natura 2000 sites under the Habitat Directive, also known as special areas of conservation (SACs). NFI measurements were conducted only for plots located on sites classified as forest areas pursuant to Polish regulations. In the study period each sample plot was measured once [23]. To determine which sample plots should be included in the study, the authors used a spatial dataset in the form of ESRI Shapefile layers including:


These input data were integrated using Qgis 2.14 software to ensure information compatibility and generate an information layer for tree stands and sample plots within the boundaries of Natura 2000 sites. As a result, 312 Natura 2000 sites (SACs) with an overall area of 3,102,247 ha, and with a total of 5557 NFI sample plots located on them, were available for the study. Those SACs included the habitat types listed in Annex I and the habitats of species listed in Annex II. The next step involved the identification of the location and type of the various Natura 2000 habitats on the aforementioned

SACs [7], as well as the NFI sample plots within their boundaries. For that purpose, we used Standard Data Forms providing information about the habitat types present on a given site, as well as conservation plans (or drafts of those plans) specifying their location. For Natura 2000 sites without protection plans, natural habitats were identified using methodological keys based on taxonomical descriptions of tree stands [26], as well as the available literature, naturalists' notes, and manuscripts concerning a given Natura 2000 site. In total, the 312 selected Natura 2000 SACs were found to contain 15 habitat types (including four priority types) with a total area of 711,306 ha. While 1620 NFI sample plots were located within the boundaries of 14 of those habitat types (see Table 1), as many as 3937 sample plots present in Natura 2000 SACs were found in habitats not listed in Annex I to the Habitat Directive [7]; in this work the latter were designated as "no habitat type" (NHT).



<sup>1</sup> Habitat type 9140 was excluded from comparative habitat analyses due to the small number of plots. \* Priority habitat types.

In addition, the Web Map Service made available by the Main Office for Surveying and Cartography was used with the C-GEO software package (with web connection) to assess elevation above sea level for each sample plot. Elevation was determined in accordance with the Polish system PL-KRON86-NH by means of interpolation algorithms prepared on the basis of the Numerical Terrain Model with a 1 × 1 m grid.

The criteria concerning deadwood volume and the density of large deadwood may differ depending on the specific characteristics of a given habitat. Those criteria are specified in manuals on habitat evaluation in Poland [10,27]. In most cases, a favorable conservation status requires a deadwood volume of >20 m3 ha−<sup>1</sup> and a density of at least 3–5 large deadwood pieces per ha; large pieces are understood as those having a diameter/DBH of >50 cm (or in some cases >30 cm) and a length/height of >3 m (Table 2). Medium-sized trees are defined as those with a diameter/DBH in the range of 30–49.9 cm, and large trees as those with a DBH of ≥ 50 cm.

In some cases, the deadwood volume threshold in a given habitat may be expressed as a proportion of stand volume rather than in absolute values. Furthermore, in some habitat types, such as 91T0, the presence of deadwood is generally undesirable given the adopted conservation priorities. This is due to the fact that large amounts of deadwood on the forest floor lead to rapid enrichment of the substrate in biogenic substances, thus increasing the competition of bryophytes and herbaceous plants to the detriment of terricolous lichens. While small amounts of deadwood are not harmful, large quantities of twigs and branches left, e.g., in the aftermath of management procedures, may cause habitat degradation [27].

#### *2.2. Data Analysis*

Data from sample plots were used to determine the mean deadwood volume for the studied habitat types. The volume of living trees and the proportion of deadwood volume to living tree volume were assessed for each habitat type to account for variation in site productivity. Then, the density and variability of medium-sized and large deadwood pieces were evaluated. Due to NFI methodology and the definition of medium-sized and large deadwood pieces (Table 2), the diameter of logs at the thicker end was calculated based on measurements taken halfway along their length (adopting a mean taper of 1 cm per 1 m). Statistical differences between habitat types were evaluated by analysis of variance and the Kruskal–Wallis test implemented in Statistica 13 software (StatSoft, Kraków, Poland).

The factors affecting deadwood volume in the studied habitat types were evaluated using a logistic regression model [28]. The choice of that statistical tool was dictated by the uneven distribution of deadwood, whose volume varied greatly among sample plots and which was absent from approx. half of them. The logistic regression model used a dichotomous dependent variable [28]. Sample plots with a deadwood volume of >20 m<sup>3</sup> ha−<sup>1</sup> were assigned the value of 1, with 0 assigned to other plots. The 20 m<sup>3</sup> ha−<sup>1</sup> threshold corresponds to the favorable conservation status as defined for most of Natura 2000 habitat types. The adopted independent variables were factors that may affect deadwood volume, such as stand, terrain, and site characteristics, as well as protection type, also obtained from the NFI. The protection type variable assumed three values: active, strict, or managed forest. For the purposes of this work, 'managed forests' are defined as forest areas that are managed with no active or strict protection plans. In turn, active and strict protection plans are most often used in nature reserves and national parks. Terrain was described by two variables: elevation above sea level (m a.s.l.) and the percentage slope of sample plots (%). Tree stands were characterized by the age of the dominant tree species (years), the volume of living trees (m3 ha−1), and tree density (trees ha–1). The model also included site fertility (dystrophic, oligotrophic, mesotrophic, eutrophic) and moisture (mesic, moist, boggy), as those parameters varied considerably among the studied habitat types. Another independent variable was habitat type, operationalized by assigning one of 13 Natura 2000 habitat codes, or "no habitat type" (NHT) for sample plots in habitats not included in Natura 2000.



<sup>1</sup> The diameter threshold was lowered to 30 cm in habitats where trees do not normally reach 50 cm for natural reasons. Those values represent diameter at breast height (DBH) for standing trees and either DBH (if measurable) or diameter at the thicker end for downed deadwood. <sup>2</sup> Diameter threshold lowered to 30 cm. <sup>3</sup> This value refers exclusively to downed deadwood in relation to stand volume, e.g., <5% means that downed deadwood volume amounts to less than 5% of stand volume. <sup>4</sup> In relation to stand volume. <sup>5</sup> d.n.a.—not available.

The model was built using the step-wise forward method (a model constructed using the step-wise backward method arrived at the same set of significant variables). The odds ratio was calculated to characterize the effects of independent variables on the dependent variable. In the case of quantitative independent variables, an increase or decrease by one unit increased or decreased the probability for the dependent variable to assume the value of 1 by the odds ratio. In the case of qualitative variables, the odds ratio was adopted in the form of reference values for each of them. The independent variables

were tested for intercorrelations. The quality of the model was evaluated by means of Nagelkerke values and the Hosmer–Lemeshow test [31]. A successful classification test was carried out on the basis of observations used to estimate the parameters of the model [32].

#### **3. Results**

The mean deadwood volume for the entire Natura 2000 area was 12.7 m3 ha<sup>−</sup>1, with very large differences between habitat types (Kruskal–Wallis H = 235.7; *p* < 0.05). The lowest deadwood volume was found for Central European lichen Scots pine forests (91TO–2.5 m3 ha−1) and Old acidophilous oak woods (9190–4.4 m3 ha−1), and the highest for Riparian mixed forests (91F0–43.1 m3 ha−1) and Acidophilous *Picea* forests of the montane to alpine levels (9410–55.4 m<sup>3</sup> ha−1). Furthermore, substantial variation was recorded for sample plots located within one habitat type, as reflected by very high standard error values (Table 3).

**Table 3.** Mean deadwood and living tree volumes and their ratio for individual habitat types.


**\*** values with different letters differ significantly at *p* < 0.05 as evaluated by the nonparametric Kruskal–Wallis test with a post hoc correction for the number of comparisons).

The mean volume of living trees for the entire Natura 2000 area was 305 m3 ha−1. Among the habitat types, the lowest values were found for bog woodland (91D0–238 m3 ha−1) and Central European lichen Scots pine forests (91T0–282 m3 ha<sup>−</sup>1). Volumes in the range of 300–400 m3 ha−<sup>1</sup> were recorded for seven habitat types, with the highest value for Sub-Atlantic and medio-European oak or oak-hornbeam forests of the *Carpinion betuli* (9160–478 m<sup>3</sup> ha−1), with differences between habitats often reaching statistical significance (Kruskal–Wallis H = 301.7; *p* < 0.05, see Table 3). The ratio of deadwood volume to stand volume ranged from approx. 1% in 91TO and 9190 to 11.4% in 91F0 and 17.0% in 9410 (Table 3).

Medium-sized and large living trees (DBH ≥30 cm) were found in all habitat types, ranging from 94 trees ha−<sup>1</sup> in 91D0 to 164 trees ha−<sup>1</sup> in 91I0, with a mean value of 121 trees ha−<sup>1</sup> (Kruskal–Wallis H = 147.3; *p* < 0.05) (Figure 2). Medium-sized and large deadwood was not found in Euro-Siberian steppic woods with *Quercus* spp. (91I0) and in Central European lichen Scots pine forests (91T0). In the other habitat types its mean density ranged from 2.5 pieces ha−<sup>1</sup> (9190) and 2.7 pieces ha−<sup>1</sup> (91D0) to 16.3 pieces ha−<sup>1</sup> (9130) and 31.1 pieces ha−<sup>1</sup> (9410) (Figure 3). Differences were also found when analyzing the density of large living trees only (DBH of ≥50 cm, Kruskal–Wallis H = 878.1; *p* < 0.05), which ranged from 4 trees ha−<sup>1</sup> for 91T0 to 61 trees ha−<sup>1</sup> for 9180, with a mean of 16 trees ha−<sup>1</sup> (Figure 2). Large deadwood pieces were absent in a total of five habitat types. In the other habitat types their mean density ranged from 0.4 pieces ha−<sup>1</sup> (9170) and 0.8 pieces ha−<sup>1</sup> (91E0) to 5.6 pieces ha−<sup>1</sup> (91F0) and 6.1 pieces ha−<sup>1</sup> (9410) (Figure 3).

**Figure 2.** Density of medium-sized and large living trees. Values with different letters differ significantly at *p* < 0.05 as evaluated by the nonparametric Kruskal–Wallis test with a post hoc correction for the number of comparisons. Uppercase letters refer only to trees with DBH ≥ 50 cm while lowercase letters refer to all trees with DBH ≥ 30 cm. Note: DBH–diameter at breast height.

**Figure 3.** Characteristics of medium-sized and large deadwood in the various habitat types.

The inclusion of factors other than habitat type in the analysis (Table 4) substantially changes the picture of deadwood accumulation compared to analysis based exclusively on habitat types (Table 3). The logistic regression model indicates that deadwood volume is mostly determined by factors associated with terrain and site accessibility, type of protection, as well as soil and stand parameters (Table 4). From among the nine variables entered in the model, the slope of sample plots and living tree density failed to reach statistical significance, while elevation above sea level, protection type, site fertility and moisture (water abundance), the age of the dominant tree species, and the volume of living trees were significant. An increase in the quantitative variables (elevation, stand age, living tree volume) was associated with an increase in the odds ratio, or the likelihood of finding a favorable deadwood volume (>20 m3 ha−1) in a given habitat type. In terms of site conditions, the reference value corresponded to sites poorest in nutrients. The odds ratio increased with both site fertility and moisture; in the latter case the highest odds ratio was recorded for boggy sites. There was a substantial difference in the odds ratio between managed forests and areas subjected to either active or strict protection. A significant effect was also found for habitat type, which was the last element entered into the model. No habitat type exhibited a significant difference as compared to the reference value in the model (Central European lichen Scots pine forests–91T0).

**Table 4.** Logistic analysis results for the likelihood of a sample plot exhibiting a deadwood volume greater than 20 m3 ha−<sup>1</sup> (the threshold value for a favorable conservation status in most habit types).


Quality characteristics of the model: Likelihood-ratio test: χ<sup>2</sup> = 939.5; *p* < 0.0001; Nagelkerke's coefficient *R*<sup>2</sup> = 0.272. The model correctly predicted results in 85% of the cases (44% for 1 and 92% for 0); the Hosmer–Lemeshow test = 13.4; *p* = 0.10. \* All the *p*-values are from Wald's tests.

#### **4. Discussion and Conclusions**

It is crucial to develop appropriate strategies for Natura 2000 sites to aid policymakers and managers in reaching biodiversity targets [33,34]. Despite an increase in Europe's afforestation, it is estimated that only 15% of its woodland qualifies for a favorable conservation status [35]. The study provides a general overview of the conservation status of forest habitats in Poland. The use of

a large number of sample plots made it possible to determine mean values for 13 habitat types. However, it should be noted that the status of a given habitat type may vary between different SACs. The statistical method applied by the NFI, employing a random, evenly distributed network of sample plots, precludes the evaluation of individual SACs due to an insufficient number of representative sample plots in each of them. Nevertheless, it is an excellent, objective tool providing a general characterization of Natura 2000 sites. In addition, logistic regression analysis revealed the site and stand characteristics that have a positive or negative effect on deadwood accumulation in areas designated under the Habitats Directive. Knowing the characteristics of individual SACs and the factors conducive to deadwood accumulation, one can predict deadwood volume for the various areas.

In Poland, deadwood thresholds adopted for most habitat types are 20 m3 ha−<sup>1</sup> for favorable conservation status and 10–20 m3 ha−<sup>1</sup> for unfavorable-inadequate status. While this is supported by some publications, those thresholds represent the lower limits of deadwood ranges proposed for European forests. Indeed, papers on the conservation of various saproxylic species or groups of species tend to suggest thresholds of 30–50 m3 ha<sup>−</sup>1, or even more [17,36,37]. In addition to the quantitative criterion, it is also necessary to ensure variability in deadwood types [38] as well as an adequate spatial distribution of deadwood microsites [16,39]. In some cases, a deadwood volume threshold may be expressed in terms of its proportion relative to stand volume. In the present study, the deadwood volume threshold of 20 m3 ha−<sup>1</sup> amounted to only a few percent of the mean stand volume (approx. 300 m<sup>3</sup> ha<sup>−</sup>1).

National parks and nature reserves, which are almost exclusively subjected to strict or active protection, revealed a markedly higher likelihood of reaching the threshold deadwood volume. However, the overall area of parks and reserves is relatively small (approximately 4% of the afforested area of Poland) as compared to that of managed forests. Given the well-established differences between managed and unmanaged woodland [40–42], it is little wonder that favorable volumes of deadwood as defined under Natura 2000 are usually found in the latter. In turn, in managed forests, deadwood volume mostly depends on the adopted management principles and their implementation. Taking into account the specific features of a given site, management procedures are determined by the species composition of the stand, its functions, as well as management objectives [43]. The implementation of different felling systems, management interventions, and regeneration patterns may result in significant differences in deadwood volume between sites [44–46]. In the present study, the average deadwood volume on Natura 2000 sites was 12.7 m<sup>3</sup> ha<sup>−</sup>1, which is more than twice higher than the mean volume reported for all Polish forests (5.9 m3 ha−1, NFI 2014). This is attributable to the fact that the Natura 2000 network primarily encompasses the best preserved woodlands in the country, including protected areas. A general assessment of Natura 2000 sites (not only forests) conducted in the years 2017–2018 as part of a periodic monitoring program revealed a declining proportion of sites with a favorable status and an increase in unfavorable-inadequate and unfavorable-bad sites [47]. In the case of some forest habitats (e.g., 91F0), general conservation status deteriorated substantially due to adverse quantitative and qualitative changes in the floristic composition, the presence of alien species, as well as hydrological disturbances [47,48]. Furthermore, it has been reported that the conservation status of many sites has been affected by excessive deadwood removal; of particular concern is the scarcity of large deadwood pieces [47].

While the management difficulty indices calculated for Polish montane and lowland forests are highly varied [49], the terrain factor was found to be significant in the model, suggesting that the higher the site elevation the higher the likelihood of finding more deadwood. A large proportion of Natura 2000 woodland sites are located in mountainous areas. Habitat types that are in part or in their entirety represented by such sites and those which are otherwise associated with steep slopes (9130, 9180, 9410) exhibited higher deadwood volumes. Additionally, monitoring reports have indicated a much better quality of forest habitats in the Alpine biogeographical region as compared to the continental region. While the conservation status of mountainous sites is usually classified as favorable or unfavorable-inadequate, that of sites in the continental region is more often deemed

unfavorable-inadequate or unfavorable-bad [47]. In managed forests, higher deadwood accumulation is significantly promoted by harvesting and skidding difficulty as well as by a less dense road network [50,51], entailing higher operating costs. Site accessibility also plays a role in lowland areas, but probably to a lesser extent [52].

While the mean deadwood volume varied considerably between different habitats, terrain and stand characteristics were of primary importance. The presented model indicated a significant contribution of stand age: the older the stand, the higher the likelihood of the site reaching the deadwood volume threshold. Since the Natura 2000 network has a relatively short history in Poland, the age structure of stands at the time of their inclusion continues to play a major role in habitat evaluation. Indeed, in the case of some habitat types this may partially explain the low deadwood volume and density of large dead trees. A good case in point are boggy coniferous forests, which were often represented by young stands at the time of their inclusion in Natura 2000, and so they have not had the time to accumulate enough large deadwood [27]. Nevertheless, stand structure analysis indicates quite high current mean densities of living trees with DBH ≥30 cm for all habitat types. Although boggy coniferous forests still reveal lower values, in the coming years they should add more large deadwood as long as they are appropriately managed. Moreover, although trees with DBH ≥50 cm are found in all habitat types, their distribution is much more irregular than that of medium-sized trees. This may be attributable to many factors, such as the adopted rotation period in managed forests or site conditions that determine the growth capacity of trees in individual habitat types. Large deadwood, which is particularly important for supporting biodiversity, is scarce or absent in many habitats. Since deadwood is deemed a crucial structural forest indicator [12], an improvement in that parameter is a crucial target that should be pursued with a view to enhancing the quality of Natura 2000 forest habitats.

In the present study, the habitat type with the greatest mean deadwood volume was Acidophilous *Picea* forests of the montane to alpine levels, although the high standard error points to an irregular distribution pattern with local aggregations attributable to frequent biotic and abiotic disturbances [53]. It should be noted that disturbances fulfill an important role in biodiversity promotion as long as the dying and dead trees are retained in the ecosystem, which is often the case in this habitat type due to its location in poorly accessible mountainous regions within the boundaries of Polish national parks. The diverse array of niches afforded by deadwood provide suitable microhabitats, shelter, and nutrition for a variety of species, increasing their numbers in a given area [54,55]. Other habitat types also contain trees which currently tend to exhibit high mortality, such as *Fraxinus excelsior* L., which occurs as an accompanying species in habitat types 91E0 and 91F0 [52,56].

The above notwithstanding, it should be noted that deadwood is not desirable in some habitats (91T0 and 91I0) as it may interfere with the conservation of the priority species occurring in them [27] due to the chemical properties of decomposing wood and its role in nutrient cycling and soil forming processes [57–59]. Indeed, both in 91T0 and 91I0 the mean deadwood volume is among the lowest with no medium- or large-sized deadwood despite the presence of large trees. However, a decision to protect certain species (e.g., rare lichens) by removing deadwood from a given habitat to prevent site eutrophication should be compensated on other sites as some saproxylic organisms have very specific requirements concerning deadwood type and other site conditions, such as insolation [60,61].

The applied regression model indicates that, in addition to protection type, deadwood volume is mostly influenced by terrain conditions, site fertility and moisture, stand age, and living tree volume. Analysis involving a dichotomous dependent variable for deadwood volume with a threshold value of 20 m3 ha−<sup>1</sup> shows that appropriate deadwood management should mitigate the effects of the aforementioned independent variables, or at least decrease their odds ratio. The factors that were found significant in the model were generally attributable to the "forces of nature." No sizable effects of management interventions were found for readily accessible terrain and for sites characterized by low growing stock. Thus, it is necessary to design a strategy for those habitats where deadwood is desirable and where standard management procedures and natural disturbances are insufficient to ensure

favorable conservation outcomes. Particularly problematic is the scarcity of large deadwood. Therefore, in managed forests fragments of saw timber stands should be left to die naturally and decay. Further monitoring is necessary as the evaluation of the Natura 2000 network depends both on its duration in individual Member States and on the adopted conservation principles for the included areas.

**Author Contributions:** Conceptualization: L.B.; methodology: L.B. and S.Z; software: L.B. and S.Z.; formal analysis: L.B.; investigation: L.B., S.Z., and M.B.; writing—original draft preparation: L.B., S.Z., and M.B; writing—review and editing: L.B. and M.B.; visualization: L.B., S.Z., and M.B. All authors have read and agreed to the published version of the manuscript.

**Funding:** This study was financed by the Ministry of Science and Higher Education of the Republic of Poland.

**Acknowledgments:** The authors thank the State Forests National Forest Holding for making the data available, as well as the anonymous reviewers for their comments.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

*Article*
