1. Introduction
Ecosystems provide a wide range of valuable goods and services that contribute to supporting nature and human well-being [
1]. These goods and services that are commonly known as ecosystem services (ES) are categorized into provisioning (e.g., marketable goods), supporting (e.g., nutrient cycling), regulating (e.g., water and soil regulation), and cultural services (e.g., recreational and aesthetic values). These services maintain the ecological processes and functions and provide resources to support the life of all organisms. Depending upon the type and conditions, ecosystems deliver unique sets of services with varying quality and quantity. For instance, a forest ecosystem provides a different set of services than grassland or aquatic ecosystem [
2], and an intact forest provides different ecosystem services to that of degraded forests [
3]. The important contribution of recognizing ES is that it reframes the relationships between humans and nature to support biodiversity conservation, ecosystem management, and sustainable development [
4].
Despite the incredible contribution of ES to support nature and human well-being, ecosystem services globally are experiencing continuous loss due to population growth, urbanization, and expansion of human settlements and agriculture [
1]. As the provisioning of ES is directly related to the type of ecosystems, e.g., land use and land cover (LULC) type in a given area, changes in LULC can cause changes in ES [
2,
5,
6]. Thus, unprecedented LULC changes in many parts of the world as a result of natural and anthropogenic activities have resulted in adverse impacts on biodiversity and ecosystems, thereby affecting their ability to provide ES [
1,
7]. Globally, LULC change has been identified as one of the significant drivers of ES loss [
8,
9]. Thus, understanding how LULC changes affect ecosystem service values is gaining momentum as a framework to communicate the values and benefits of appropriate land management to policymakers and land planners and managers [
10].
Since the Millennium Ecosystem Assessment in 2005, ES research has gained momentum and the science and policy of valuing ES has experienced rapid growth. The Economics of Ecosystems and Biodiversity (TEEB) initiative (2010) and the founding of Intergovernmental Science-Policy Platform on Biodiversity and Ecosystem Services (IPBES) (2012) are prominent examples at the global scale. The economic valuation of ES is becoming a regular practice in ecological economics to improve the basis for political and planning decisions and create a more profound knowledge on the economic importance of ES. This includes, for instance, the assessment of opportunity costs of land degradation or restoration and developing policy tools and incentive mechanisms such as payments for ecosystem services (PES) [
2,
11]. However, monetary valuation of ES does not imply that they are the exchange values and that all ES cannot be commodified or exchanged in the market as a perfect substitute to other market commodities [
8,
12,
13]. Thus, the valuation is instead an estimate of the benefits of ES to society expressed in units that work as a tool to help raise the awareness on ES importance [
8].
The application of remote sensing and GIS tools to map land use and land cover changes have opened new areas for ecosystem service research. Further, methods to quantify ES has also progressed over the years, resulting in a wide range of available methods to value ES [
14,
15]. Although inconsistencies in the methods create a challenge to include robust values of ES in national accounts, policy making, and decision making, attempts to quantify ES values are ongoing and this contributes to improving our knowledge and experience in the science of ES valuation. In this study, we used remote sensing and geographic information techniques to determine land use and land cover of two different periods and associated ecosystem services values (ESV) with a benefit transfer method using global and local databases. Our study was conducted against the background of a rapidly changing landscape in the southern plains of Nepal; it is the first total economic valuation of ES in the region.
Although Nepal has significantly improved on biodiversity and ecosystem conservation, implementation of ecosystem services concept is limited in Nepal (e.g., only independent project-based PES schemes that focus mainly on water-related services). In recent years, the Government of Nepal has taken several steps towards broader application of ES concept. For instance, the recognition of the economic value of ES in the current national development plan and the preparation of a national policy on Payment of Ecosystem Services (PES). However, integration of the full range of ES into policy and planning for conservation and management of natural resources is still lacking, the result of which is significant loss and trade-offs in ES. This study aimed to analyze land use change, and its impact on ES in Terai Arc Landscape, a program implemented in 2001 to conserve biodiversity and ecosystems and support sustainable livelihoods in the southern plains of Nepal. The study output can be valuable knowledge to inform stakeholders about ES changes and provide a base for a fine scale local level estimation in the future.
4. Discussion
In this study, we observed the loss of forests, agricultural lands, and water bodies, and a corresponding increase in shrublands and urban areas between 2001 and 2016. The loss of forests and agricultural land in the plains of Nepal conformed with several local and global land change studies [
16,
31,
32]. We found that LULC change in TAL was dynamic and non-linear, meaning that the conversions did not follow similar patterns. The spatial patterns of change showed that, while the formerly cultivated lands were converted to urban areas, the forests in the fringes of the new urban areas were converted to agricultural lands. This implies that rapid urbanization in the landscape caused not only loss of agricultural lands but also a compounding effect on the intact forests due to feedbacks from food insecurity and agricultural incursions. Although TAL program may have contributed in slowing the rate of deforestation and degradation of the tropical forests in the region, forests continue to be converted to shrublands and cultivated lands. Along with anthropogenic factors such as forest encroachment, livestock grazing, and illegal logging, natural factors such as forest fires, droughts, and invasive species continue to threaten natural ecosystems [
30] and may have contributed to forest loss and degradation and depletion of water resources in the region.
Forests, the dominant land cover (44% area of TAL), provide a wide variety and high value of ES (USD 871 million ha
−1 year
−1). However, the total ES values of forests declined between 2001 and 2016, primarily due to forest loss. The decline in ES primarily due to forest loss is consistent with the findings of other studies. In China, the loss of forests was the major contributor to the decline in ES values during the period 1988–2006 [
28]. Similarly, the loss of natural forests in Ethiopia has resulted in the decline of values from several ecosystem services [
26,
33]. The water bodies in the landscape provide a wide range of goods and services that are important for maintaining ecosystems and supporting local income generation. In terms of ES values, the waterbodies provide the highest ES values per ha (USD 2699 million ha
−1 year
−1). As the ES values decreased substantially due to loss of water resources, their conservation and management can play an important role in conserving and enhancing ES in the region. The Terai region of Nepal is also the food basket of the country and the conversion of prime agricultural lands to urban areas in TAL region can have both local and national implications on food security and socio-economic well-being. As rapid population growth continues in the landscape (that will create higher ES demand), the government should promote effective utilization and management of land to supply and balance ES in the landscape to support the well-being of both nature and people.
As we used simple, less data-intensive methods, there are some limitations of this study. First is the accuracy of land use classification. Although classification accuracies of 86% for 2001 and 88% for 2016 were achieved (which are considered high accuracy in classification techniques), there may still be some minor miscalculations that could have affected the ESV changes. Second is the use of the benefit transfer method that assumes values are constant over the type of ecosystems. However, in reality, most ecosystems are different, and the service beneficiary population vary greatly. In addition, the reliability of the benefit transfer method was not checked. Other similar studies check reliability by computing the Coefficient of Sensitivity (CS) using the standard economic concept of elasticity and adjust the Value Coefficients (VC) by ±50% [
26,
28,
33]. This method considers the values of CS (greater than 1 and less than 1) to determine if the estimated ES value is elastic (greater than 1) or inelastic (less than 1) with respect to that coefficient. However, the same results were acquired while adjusting VC to ±25%. Therefore, as the estimation is independent of the ±50% change and always gives values less than 1, this method does not adequately address the reliability, but rather suggests that the coefficients are robust [
34]. Hence, we did not use this method for the reliability check. Thus, the results should be taken as an initial crude estimate. We suggest that another robust method should be applied to generate precise and accurate values.
Nevertheless, our study added to the dimension of local level estimation of ESV by utilizing freely available Landsat images and valuation data from TEEB database that resembled local geographical settings, rather than relying only on global values from Costanza, which could otherwise underestimate or overestimate the economic values of the services. Although similar techniques have generated local values for other regions of the world, e.g., in Africa, studies of this kind are scarce in Asia. Thus, the output of our study will help the local level estimation for other data scarce regions of the world, particularly in the tropical countries of Asia. The estimation of ESV using readily and freely available remote sensing and value data is important in developing countries such as Nepal where historical land use data are not available, and ground data collection is expensive and time-consuming. This valuation will also set a stage for, and facilitate further detailed analysis for precise and accurate information by utilizing extensive data and techniques in the future. The quantitative evaluation of total and individual ESV changes as shown in this study can provide an important avenue for effective land use management decisions for sustaining land resources and optimizing the delivery of important ecosystem services.
5. Conclusions
The estimation of ESV based on LULC changes from different periods is an important indicator of how much services have changed on spatial and temporal scales in the landscape. Our study shows the shifts in the ES values of total and individual ecosystem services in response to LULC changes. We found that the expansion of urban and shrublands at the expense of agricultural lands, forests and water ecosystems during the period of 2001–2016 reduced the total ESV by USD 11 million. As forests represent by far the most significant contributor to ESV and water bodies as high ESV per ha, shifting the current LULC changes by protecting forests and preserving water resources is essential to prevent further loss of ESV. Besides, the loss of agricultural lands shows a growing threat to food security in the rapidly populating landscape. Thus, it is vital to maintaining an appropriate proportion of LULC in a landscape to provide optimal ecosystem services. The valuation of ecosystem services in this study is an attempt to communicate the benefits of ecosystems to both the national stakeholders and the broad audience and open up avenues for further detailed analysis. There are various detailed direct and indirect methods to generate higher accuracy and precision of values but they require time and resources. As our study is a first initial crude estimate based on a simple benefit transfer method, we recommend such a detailed analysis in the future. In addition, the results could be extended by developing future scenarios that consider the population growth and urbanization trend in the landscape and assess their impacts on ES. In addition, analysis of land change before TAL was established could also be useful to assess the effectiveness of the TAL program in ES provision. The management of agricultural lands, forests, and water resources are closely linked and require innovative policy solutions. As our study compared values from different ecosystems, we believe our study will serve as important knowledge for future research and policy formulation in Nepal. We also see that there are many prospects for the government to facilitate Payment for Ecosystem Services (PES) as a strategy to protect ecosystems and improve local livelihoods in the landscape.