1. Introduction
Globally, forcibly displaced population reached remarkably high from 43.3 m in 2009 to 70.8 m in 2018, largely because of persecution, conflict, violence or human rights violations (UNHCR, 2019). Presence of displaced people may introduce a range of effects to the host environment of which deforestation, land degradation and impacts on water supply and quality are noteworthy (Black, 1994). However, severe pressure on local environment causes forest cover to deplete substantially (Hagenlocher et al. 2012; Birendra and Nagata, 2006) due to enhanced competition for natural resources between displaced communities and local population (Chambers, 1986). Furthermore, clearing of forest cover evidently reduces ecosystem functions and services (Foley et al. 2007). For instance, biomass and carbon stock depletion in relation to deforestation caused by increased human activities can accelerate climatic change at regional and local scales (Panja, 2020; Bonan, 2008; Malhi et al. 2002).
Following severe violence in Myanmar’s Rakhine State in 2017, nearly a million forcibly displaced Rohingya population (a minority ethnic group) took refuge in Cox’s Bazar Teknaf peninsula of Bangladesh (UNHCR, 2017), which is believed to be the largest refugee camp in the world at present (Kolstad, 2018; IOM, 2017). Although Rohingya issue dated back in 1942 (Human Rights Watch, 2000), their influx is multiplied since August 2017 (UNOCHA, 2018). A total of 932,940 Rohingya are now being hosted in 48 temporary camps which are located either within the reserve forests or in a close proximity to forested lands (UNDP Bangladesh and UN WOMEN Bangladesh, 2018). Since Cox’s Bazar Teknaf peninsula is an ecologically critical area (ECA) (DOE, 2015), hosting such a big number of refugee population is not only a significant threat to the environment (Hassan et al. 2018; Rahman, 2017) but also has implication for local and regional security (Bashar, 2018; Rahman, 2010). Forest clearance for building camps and ensuring livelihood activities (Lynch, 2002; Rashid et al. 2020), especially within 15 km of the camps (Sato et al., 2000), are two major issues that the peninsula is currently facing. Although the Rohingya communities are supported by different organizations, these supports do not guarantee cash flow, hence they have no or limited livelihood options (IOM and FAO, 2017). Therefore, they depend on available forest resources to maintain their livelihoods (Wilson, 1994; GTZ, 1994), which is posing a significant threat to forest resources (Ghimere, 1996). Besides, cooking requires fuelwood (IOM and FAO, 2017) which is being met up through harvesting from locally available natural and community forests. Since refugees use natural resources in a more unsustainable way than local communities (Black and Sessay, 1998), long term impacts on natural resources could be irreversible.
Various impacts associated with forced displacement on host environment are noted in a number of studies (see Hagenlocher et al. 2012; Black, 1994), however the extent of ecosystem degradation depends on factors like destination country’s capacity (Hugo, 1996), degree of disturbance to forests (Shukla et al. 2011) and institutional arrangement (Black and Sessay, 1998). Since Bangladesh is densely populated and have limited natural resources, particularly forested lands, sheltering nearly one million Rohingya over the peninsula resulted in a variety of impacts, including severe depletion of forest covers (Ahmed et al. 2019). As best management practice requires information about spatiotemporal pattern of the past, present and future scenarios of a geographic phenomenon (e.g., forest ecosystem), monitoring and predicting forest extent in the peninsula with satellite data could be of great value (Bjorgo, 2000a; Lodhi et al. 1998). Although geospatial data are valuable source of information for crisis and disaster management, relatively little is done in elucidating degradation of natural forest and related ecosystem function in response to a sudden humanitarian crisis such as Rohingya. This work aims to fill this void. Despite a couple of works attempted to estimate forest loss (Ahmed et al. 2019; Hassan et al., 2018), they are however limited in data and scope. Notably, they are unable to show factors accountable for forest degradation, prediction of forest degradation (assuming business as usual scenario), and most importantly, loss of ecosystem function over space and time. Planning and efficient management of meagre natural resources requires up to date data, therefore, availability of existing forest resources and potential loss in relation to refugee migration is expected to provide crucial information to managers and policymakers (Morales-Hidalgo, 2015; Romijn et al., 2015; Bouchardy, 1995; Bjørgo, 1999). This work, thus, attempts to answer two research questions: (i) what is the extent of forest degradation in environmentally sensitive Cox’s Bazar Teknaf peninsula in response to Rohingya influx; and (ii) what is (would) be the degree of ecosystem function loss, at present and in the future, under multiple stressors and stimulus, if Rohingya issue persists.