*2.1. The Lau*

Lau Lagoon extends for approximately 35 km on the northeast coast of Malaita. The shallow lagoon harbors a rich diversity of coral reefs, seagrass meadows, and mangrove forests. There are approximately 94 artificial islands in the lagoon [23]. Several more artificial islands are located in neighboring Suava Bay. These settlements are built by manually hauling and piling up coral rocks on shallow reefs [24]. Small new extensions are constructed for new households. As a result, some islands form a maze of small raised platforms connected by narrow bridges. Some of these man-made islands, such as Sulufou, Funafou, Foueda, and Tauba, are relatively large (>1 ha), and are densely populated. Others are very small (<100 m2). Some islands are built as extensions of natural islands or rock outcrops in the lagoon. Others are constructed in the mangroves by constructing coral rock walls, often more than 3 m high, and filling the enclosure with gravel and sand. In most cases, these islets are just above the high-water mark (<30 cm), and most houses are constructed on stilts (see Figure 3).

**Figure 3.** Funafou Island in Lau Lagoon (J. van der Ploeg, 2015).

Fishing forms the basis of people's livelihoods in the lagoon, and the Lau have an in-depth knowledge of their marine environment [25]. A great variety of fishing methods is used. Fish, crabs, marine turtles, and a variety of shells are bartered for root crops and vegetables with Baegu and Baelelea farmers from the uplands [26]. A complex tenure system regulates access to and use of marine resources in the lagoon. Coral reefs and deep pools are generally claimed by patrilineal clans, locally referred to as tribes [27]. In principle, land and sea rights are exclusionary, but in practice other clans often have usufruct rights. Seagrass meadows and the deep sea are de-facto open access. Traditionally, the Lau have managed their fisheries by imposing temporal closures for certain reefs [28].

In the 19th century, there were virtually no settlements along the coast of Malaita, with the exception of the artificial islands of the wane asi [21]. During the colonial period, wane tolo moved from their small, scattered hamlets in the uplands to large, permanent villages in the coastal areas [6]. Many wane asi left the artificial islands to settle in these new communities, a process that has continued ever since. Today, a large Lau community resides in the national capital, Honiara, and the provincial center, Auki. But the artificial islands remain important for people's identity and worldview, also for people born and raised in town. Several artificial islands in the lagoon, such as Abu, Kwaleunga, Longoaia, Kwaloai, Madanga, Kwailabesi, and Foufoiasi, have been abandoned. At the same time, new islands are constructed in the lagoon, mainly to develop tourism facilities. Nowadays, approximately 3600 people live in Lau Lagoon; much less than one hundred years ago [29].

Two Lau communities, Walande and Fanalei, are located on South Malaita, approximately 130 km south of Lau Lagoon. At least 12 generations ago, people settled here to hunt dolphins [19]. The porpoises are killed for their teeth, which are used for customary marriage and compensation payments, and for their meat, which is an important source of income and food for these communities. These two islands feature prominently in the climate change discourse in Solomon Islands as prime examples of sinking islands [30,31] (see Figure 4). Approximately 750 people now live in these two communities.

**Figure 4.** The remains of the Anglican church of Fanalei (J. van der Ploeg, 2018).

A closely affiliated ethnic group are the Kwai, who live on two densely populated islands on the East coast of Malaita: Kwai and Ngongosila. People here speak Guala'ala'a, which was used as a trading language along the coast [21]. Reliable census data is lacking but it's estimated that around 900 people live in these two communities. The saltwater people from Kwai and Ngongosila, and several other small artificial islands scattered around Uru Harbor, trade fish with the Kwaio people from the uplands. Ngongosila was settled in 1955 when the South Sea Evangelical Mission built a church on the island [29].

### *2.2. The Langalanga*

Langalanga is a 22 km long lagoon on the west coast of Malaita, and is one of the most densely populated regions of the country. Historically, the Langalanga people bartered fish for crops with the Kwara'ae, shifting cultivators who inhabited the forested hinterlands of the lagoon [32]. The saltwater people built artificial islands on the barrier reef of the lagoon and in the mangroves, and specialized in the production of tafuli'ae—strings of polished shells, which are traditional wealth items used throughout the Solomon archipelago for trade, feasts, and compensation and marriage payments [33,34]. There are around 59 artificial islands in Langalanga lagoon, most of them located in the mangroves and sago swamps. Approximately 6000 people live in the lagoon, including those in the settlements around Auki such as Aoke Island, Niu Kaloka, Ambu, and Lilisiana [29].

Historically, the livelihoods of the Langalanga people were characterized by much geographical mobility: Fishers moved along the west coast of Malaita to exploit a variety of marine resources, and they traded shell money with people from different islands [35]. European contact fundamentally altered livelihoods, trade networks, and social relations in the lagoon. The labor trade and evangelization efforts provided new goods such as steel axes, fishhooks, and guns, which enhanced agricultural productivity and led to widespread violence [21]. In 1909, the British colonial government established a station at the northern tip of the lagoon, present-day Auki town. In the 1930s, several shipyards were established in the lagoon. As a result, Langalanga people dominated inter-island trade in the British Solomon Islands Protectorate [36]. But in other ways the mobility of saltwater people, which was an important strategy to respond to environmental changes and shocks, became increasingly restricted. As more and more people settled along the coast, conflicts erupted over access to fishing grounds. Despite these developments, fishing remains the primary source of food and income for the saltwater people of Langalanga [37]. However, the widespread use of dynamite in the 1950s and overharvesting have led to a rapid decline in the productivity of the fisheries [38,39]. The limited prospects beyond the subsistence economy have stimulated urbanization: The saltwater communities around Auki town,

such as Lilisiana, have grown rapidly over the past fifty years (see Figure 5), and many young and educated people migrate to Honiara in search of a better life.

**Figure 5.** Lilisiana in Langalanga Lagoon (J. van der Ploeg 2018).

#### **3. Methodology**

To illustrate the popular sinking islands narrative, a qualitative content analysis was conducted on articles published in the Solomon Star from January 2015 to March 2020 on the effects of climate change in Malaita Province [40,41]. Articles on climate change that did not specifically focus on Malaita were not included. Articles in the other daily newspaper in Solomon Islands, the Island Sun, were not included because this newspaper is often unavailable on Malaita. In total, 73 articles were compiled (in print and on-line) and encoded. Newspaper articles on climate change were published irregularly, with quite a few articles published in March 2015 in the aftermath of cyclone Pam, and in September 2017 in preparation for the United Nations climate change conference in Bonn, chaired by Fiji. A coding framework was developed by deriving categories directly from these newspaper articles. A potential problem of this inductive approach is the identification of all potentially relevant categories in the absence of a theoretical model. Yet, the categories of the coding framework overlapped to a large extent with key themes from the academic literature.

This paper tries to contextualize some of the claims made in these newspaper articles and reconstruct local socio-ecological events through an iterative process of inference and induction. Bradley Walters and Andrew Vayda [42] advocate such analysis to unravel the interacting causes of environmental change. Instead of relying on preconceived conceptual models and questionnaires, this flexible methodology enables the researcher to pursue lines of enquiry that emerge during fieldwork. An example from the field can illustrate this. Coastal erosion is threatening the small island of Ta'arutona in the West Are'are Lagoon, a process that is often attributed to climate change. Instead of asking if the recurrent floods were the effect of climate change, the authors walked around the island with key informants, asked people to describe the events, and discussed plausible explanations. The sinking of Ta'arutona Island seems to have started in the year 2000, after the mangroves on the island were cut for firewood to dry copra. Villagers say that the removal of the mangrove buffer has exposed the island more directly to waves. In December, king tides overflow the island, particularly when there is a strong northwestern wind. The floods, sometimes up to 30 cm above the ground level, destroy homegardens and coconut groves, and damage houses, despite the efforts of the villagers to build sea walls. People are concerned about an impending tsunami, and several households have re-settled on the mainland. Most young people from the village have moved to Honiara, which makes the maintenance of the sea walls problematic. Climate change induced sea level rise could play a role in the flooding of the island, but other neighboring villages in the lagoon, such as Pipisu and Rohinari,

seem much less affected. As such, a more nuanced, complex, and uncertain explanation emerges for the environmental changes on Ta'arutona.

The analysis draws primarily on ethnographic fieldwork on Malaita in the period 2015–2018. In this period, the authors made several field trips in the context of the research and development program of WorldFish in the province [43]. The authors made repeated visits to villages in Langalanga Lagoon, Lau Lagoon, West Are'are, East Malaita, and South Malaita to identify threats to coastal fisheries and rural livelihoods. Spontaneous interviews with community leaders, fishers, school teachers, church elders, customary chiefs, and members of women and youth groups were held on-site using Solomon Islands Pijin, the lingua franca of the country. In total, informal interviews were conducted with 171 people (61 women and 110 men) in 62 villages (see Figure 6 for the location of the villages mentioned in the text). In accordance with the WorldFish policy on ethics of research involving people, all respondents gave verbal prior and informed consent for an interview. No cash payments were made to the respondents. Community meeting were held in all villages before conducting the interviews to explain the aim and methods of the research. Information collected during the interviews was recorded in the author's notebooks, and triangulated when possible. These interviews were complemented with information from the community ward profiles compiled by the Provincial Government Strengthening Program (PGSP) and the community profiles of the Rural Development Program, Phase 2.

**Figure 6.** Artificial islands on Malaita mentioned in the text, highlighting the main reasons why people abandon these places.

This research methodology clearly has limitations. First, by relying mainly on local ecological knowledge to describe and understand complex and long-term biophysical and ecological processes, this analytical framework is subject to criticism. But in the absence of quantitative, locality specific data, a situation that is unlikely to change in the near future, it is the only feasible way to generate empirical information in many remote rural areas in the tropics [44]. Second, this analytical framework might seem to have limited value for policy makers, donors, and development practitioners. This paper

does not provide practical recommendations to improve CCDRM projects targeting the wane asi. Instead, the aim is to nuance a priori assumptions on the local impacts of climate change, and to better understand the links between climate change and other environment and development problems. It is hoped that this will lead to a better allocation of climate change funding.
