Next Article in Journal
Renewable Energy Project as a Source of Innovation in Rural Communities: Lessons from the Periphery
Next Article in Special Issue
Factors Influencing Farmer Willingness to Fallow Winter Wheat and Ecological Compensation Standards in a Groundwater Funnel Area in Hengshui, Hebei Province, China
Previous Article in Journal
Heuristic vs. Meta-Heuristic Optimal Energy Design for an Office Building
Previous Article in Special Issue
Spatial Analysis of Chinese Grain Production for Sustainable Land Management in Plain, Hill, and Mountain Counties
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Projections of Future Land Use in Bangladesh under the Background of Baseline, Ecological Protection and Economic Development

1
Institute of Geographic Sciences and Natural Resources Research (IGSNRR), Chinese Academy of Sciences (CAS), Beijing 100101, China
2
University of Chinese Academy of Sciences, Beijing 100049, China
3
Center for Chinese Agricultural Policy, Institute of Geographic Sciences and Natural Resources Research (IGSNRR), Chinese Academy of Sciences (CAS), Beijing 100101, China
4
Department of Agricultural Extension and Rural Development, Bangabandhu Sheikh Mujibur Rahman Agricultural University (BSMRAU), Gazipur1706, Bangladesh
5
Faculty of Resources and Environmental Science, Hubei University, Wuhan 430062, Hubei, China
*
Author to whom correspondence should be addressed.
Sustainability 2017, 9(4), 505; https://doi.org/10.3390/su9040505
Submission received: 21 February 2017 / Revised: 21 March 2017 / Accepted: 22 March 2017 / Published: 28 March 2017
(This article belongs to the Special Issue Sustainable Land Use in China)

Abstract

:
Land is one of the important input resources in a highly populous and land-scarce country such as Bangladesh. When different factors change (such as, geophysical, proximity, socioeconomic and climatic), there are dramatic changes in the spatial pattern of land uses. Thus, shedding light on the dynamics of land use and land cover changes has great importance for finding the changing pattern of land use in Bangladesh. In the present study, we predicted the land use and land cover changes from 2010 to 2030 under baseline, ecological protection priority and economic growth scenarios in Bangladesh. On this basis, we applied a previously developed Dynamics of Land Systems (DLS) model to simulate the changes in land uses according to the driving mechanisms. The findings indicate that cultivated land declines and built-up area expansion is common under all three scenarios. However, the future land use demand shows differences under different scenarios. The results under the ecological protection priority scenario shows that forest area and grassland will increase more, while under economic growth scenario, built-up area will expand dramatically in the future. The present research results furnish meaningful decision-making information for planners to conserve and/or exploit land resources in Bangladesh in a more sustainable manner.

1. Introduction

Bangladesh is one of the least developed countries of the world [1], and is mostly rural and agricultural [2]. Bangladesh has a large population (144.1 million) and high population density (964/km2) [3]. Bangladesh has faced many factors in recent years that driving land use and land cover changes (LUCC) such as population dynamics; rapid changes in economic growth; climate change; construction of roads and highways; electrification; more advanced agriculture, technology and irrigation facilities; extended education; improved health services; new residential infrastructure; etc. [4]. Rapid land use and land cover change (LUCC) induced land degradation, together with climate change and human activities, is thought to be a threat worldwide [5,6,7]. Asia has also experienced impressive land cover changes, yielding land degradation, and ecosystem service and economic loss [8,9].
Land use is conceived as a manmade adjustment of the land surface, and has a crucial role in shaping ecosystem functions [10]. Land use change is the modification system of land into the cultivated land, grassland, forest, and urban land and land use and land cover change (LUCC) is one of the key factors for quantifying the natural environmental and human activity related changes [11]. In general, LUCC is termed as the modification of several land types by human beings [12,13].
In Bangladesh, few studies have been conducted thus far on LUCC. Utilizing the Remote Sensing technique during the period of 2001 to 2008, Shapla et al. classified land use types of Gazipur district and detected the increasing trend of built-up and urban areas [14]. In another study, Kumar and Ghosh utilized Remote Sensing (RS) technique to detect the land cover change on Hatya island from 1977 to 1999 and noticed that cultivated land and forestlands were increasing while grassland was decreasing [15]. Through Advance Land Observing Satellite (ALOS) technique, Zaman and Katoh identified that population pressure and overexploitation of resources helped to reduce the forest cover in Takurgaon district [16]. In another study, Rahman and Begum, through RS technique, identified, in the Sudarbans area, shrinkage of grassland and forest cover by the transformation of unused land [17]. Islam and Hassan used remotely sensed data in Rajshahi district and acknowledged that built-up area was increased by the loss of agricultural land from 1977 to 2010 [18]. Mamun et al., utilizing GIS and RS technique on Dhaka City, revealed that built-up area expanded by the transformation of agricultural lands, water areas and unused lands [19]. Roy et al. monitored and predicted land use changes in the Chittagong Hill Tracts by utilizing remote sensing technique and Markov-chain cellular automata model. They simulated the land cover change of the study area from 2014 to 2042. Although their study exhibited a variation of the land use types of the study area, they did not consider important factors such as physical, socioeconomic and policy that might affect the land use change. They recommended further studies incorporating relevant factors [20]. Ahmad et al. utilized Conversion of Land Use and its Effect (CLUE-s) model to identify the key driving factors influencing urban growth of Dhaka City [21]. They simulated urbanization for the periods 1988–1999 and 1999–2005. Islam and Ahmed utilized Markov Chain Modeling technique to detect the future land cover map of Dhaka City in 2020 and 2050 based on past trend from 1991 to 2008. Their findings suggested that built-up area would increase in the future after the conversion of cultivated lands and vegetation [22]. Through a similar study utilizing the Markov Chain Cellular Automata model, Corner et al. suggested that cultivated land and built-up area will decrease in Dhaka City during 2022–2033 [23].
The World Bank [24] defines sustainable land management as practices and technologies that target incorporating the overall management of land, water, biodiversity and other environmental resources which can fulfill the human requirements for ensuring the long-term sustainability of ecosystem services and livelihoods. Global environmental change results in soil erosion, shortage of water, drought, loss of biodiversity and ecosystem damage [25]. Xie et al. conclude that ecological land is perceived as an important natural resource and necessary for preserving environmental security [26], an essential factor to attain economic development and protection [27]. Any form of alteration of ecological land can start different types of environmental problems [26]. Furthermore, ecological land also needs to include both cities and rural areas. Yao and Xie in China recognized and explored the multiple functions of the countryside, including agricultural production, ecological services, aesthetic and environmental values. Their findings also exhibited that the government’s scientific planning and controlling of rural land resources could ensure the most economical land use, ecological welfare and food security [28]. Therefore, it is important to note that LUCC is considered a crucial factor affecting global environmental change and environmental protection [29,30,31]. Hence, sustainable land management is essential in a land-scarce country such as Bangladesh.
It is common that cultivated lands are converted into nonagricultural purposes in most countries, and the situation is more severe in Bangladesh due to huge population pressure. Due to this cultivated land lost, agricultural production has been hampered in Bangladesh. Thus, LUCC research is a good topic for scientists of Bangladesh. In Bangladesh, very few studies have been conducted so far on future land use change. Most of the conducted studies utilized the remote sensing technique to show LUCC in different regions of Bangladesh. Thus, we can assume that the use of different land use models for future land change prediction in Bangladesh is rare. Therefore, it is crucial to simulate the dynamics of land systems, which ultimately helps in land management decision-making and land use planning [32]. However, land change processes are a complex issue resulting nonlinear developments [33] and human decisions directly or indirectly affect land change process [34]. For the last 20 years, various land use models have been adopted throughout the world [35], and most are used to predict future land cover changes [36].
In this context, we deployed the Dynamics of Land System (DLS) model to simulate the dynamics of the land system by integrating multiple data sources. The DLS model is a widely practiced model of simulation of the land system [32,37,38,39,40,41]. DLS model has two special features: firstly, it provides scenario analysis of land demand at regional level and spatial integration of land uses at pixel level. Secondly, it can consider the interactions among influencing factors of land use and neighboring pixels for those influencing factors. In this study, we will examine the future land use pattern in Bangladesh and will forecast the future land use under baseline, ecological protection priority, and economic growth scenarios so Bangladesh can develop a sustainable land use pattern.
This article proceeds by a brief introduction of the study area in Bangladesh. Then, we discuss the research methodology, including variables and parameters used in the scenario analysis. This section also includes a detailed description of the three scenarios. Next, we explain the DLS model methods for predicting future condition for each of the three scenarios. Finally, we conclude through the presentation of crucial results and discussing their significance.

2. Data and Methodology

2.1. Study Area

The study area encompasses the entire Bangladesh. The country is located between the latitudes 20°34′ and 26°38′ north and the longitudes 88°01′ and 92°41′ east. Bangladesh covers lowlands of the Indo-Gangetic Plain, which is terminated by the two main rivers (the Ganges and the Brahmaputra) with alluvial plain. Tertiary hill regions are found as a part of the Himalayan range. Two hill regions are seen, Chittagong hills in the southeast and low hills in the northeast and modest-elevation highlands in the north and northwest. Southwestern Bangladesh is occupied by the Sundarbans, which is characterized by marshes and swamps. North and west Bangladesh is boarded by India, while the Bay of Bengal is in the south of Bangladesh. Both India and Myanmar are east of Bangladesh. Total area of the country is 147,570 km2. The territorial water area is 200 nautical miles. Administratively, there are seven divisions (as of 31 December 2015, there are eight) in the country: Dhaka, Chittagong, Khulna, Rajshahi, Barisal, Sylhet and Rangpur. The newly added division is Mymensingh. There are 64 Districts (Figure 1), which are key administrative units of the country. Total number of Upazilla (the lowest administrative unit) is 489. The total population of Bangladesh is 144.1 million, of which the urban population is 33.5 million and rural population is 110.6 million [3]. Annual growth rate of population is 1.37%. Bangladesh is one of the most densely populated countries in the world and her population density is 1062.5 per km2 [42].

2.2. Data

2.2.1. Land Use Data

Land use data from 2000 and 2010 for Bangladesh, with a resolution of 1km, were used in this study. The original land use maps (map of 2000 and 2010) were derived from www.globallandcover.com [43]. These are 30m fine resolution global land cover maps developed by China Globe Cover [44]. Primarily from the global land cover map, four tiles contain the Bangladesh land cover map for each year: for 2010, N45_20_2010LC030, N45_25_2010LC030, N46_20_2010LC030, and N46_25_2010LC030; and, for 2000, N45_20_2000LC030, N45_25_2000LC030, N46_20_2000LC030, and N46_25_2000LC030. After downloading the tiles from the website, they were processed through mosaicking using the software ArcGIS 10.1. Then, the whole Bangladesh land cover maps are derived by clipping through ArcGIS 10.1 software (ESRI, Redlands, CA, USA).
According to the requirements of the DLS model, it is necessary to prepare all the data with unified projection and spatial registration. Moreover, most of the driving factor data were premade in 1 km × 1 km resolution grid data. Thus, after extracting the land use data, we transformed 30 m resolution to a grid format with spatial resolution of 1 km × 1 km in the same projection coordinates [37,45]. Then, the land uses are reclassified into six types [32,40,41,46,47] with different coding according to Table 1 and are generalized through a distinct representation scheme. Meanwhile, the land use data from2010 were adopted to assess the accuracy of the simulation with the DLS model.

2.2.2. Different Influencing Factors Data of Land Use Changes in Bangladesh

Different factors influenced land use pattern [41,48]. We utilized four categories of influencing factors during this study: geophysical, climatic, proximity and socioeconomic factors. These variable and factors are explained in Table 2.
The land use types and factors influencing land use are used for simulating the land use scenarios in this study. Then, we identified, measured and considered the relationships between the binary variable for each land use and their influencing factors through regression analysis.
First, we collected all of the factors and processed at 1 km × 1 km pixel level resolution [37]. Out of these factors, the climatic data were collected from the Bangladesh Agricultural Research Council (BARC). There are 35 weather stations present in Bangladesh. Monthly average climatic baseline data for all 35 weather stations were collected for this study. Then, the climate data were interpolated by means of kriging method of ArcGIS 10.1 [32,41,51]. We collected geophysical data from the International Soil Reference and Information Center (ISRIC) World Soil Information Service [52,53], and Ministry of Agriculture (MoA), Bangladesh. The data were pre-made at 1 km × 1 km grid level [32]. The estimation of proximity data was done using the national highway and divisional data with the ArcGIS software. The socioeconomic dataset was derived from the Statistical Year Book of Bangladesh [54], and the population data were collected from The World Population Organization [55]. Finally, logistic regression analysis was employed to find the statistical relationship between the quantitative changes and spatial dynamics of land use changes, their driving factors, and the weight coefficient matrix of the land use change driving factors [56,57]. We carried out the logistic regression by means of SPSS software. The logistic regression formula was as flows:
y = β0 + β1x1 + β2x2 + … + βkxk + ε
where, y is the probability of existence of one specific land type in the cells under given driving factors; X1, X2, …, Xn indicate the driving factors of different categories such as location, climate, population, GDP etc.; and β1, β2, …, βn are coefficients of regression analysis of driving factors. β0 is constant.

2.3. Methodology

2.3.1. Land Use Scenarios

There exists a convenient relationship between scenario analysis and land use projections for each year of the simulation period. We developed scenarios for the land use and land cover changes in Bangladesh for the present study based on a literature review related to future Bangladeshi plans and expert judgments. Before developing scenarios, we reviewed several literature documents, including the Prospective Plan of Bangladesh 2010–2021: Making a vision 2021 a reality [58]; 6th Five Year Plan (FY2011–FY2015) of Bangladesh [59]; 7th Five Year Plan (FY2016–FY2020) of Bangladesh [60]; Statistical Year Book of Bangladesh [42]; and the World Bank Data [61]. We also consulted experts in developing the scenarios, both the Chinese and Bangladeshi parts. Based on literature review, expert opinion, and the characteristics of land use and regional development of Bangladesh, we embodied three scenarios, baseline, ecological protection priority, and economic growth, which were used in the Dynamics of Land Systems Model (DLS) to simulate the changes of land systems of Bangladesh.
In this study, taking the land use data of the year 2010 as baseline, we simulated the land use in 2030 under the three scenarios.

Baseline Scenario

The Baseline scenario is a reference case scenario delineating a future condition of society and environment where no new economic and environment policies are implemented, and the same population growth and economic growth trend continue. The baseline scenario is constructed according to the previous statistical data (e.g., 2000 and 2010 for the current study) and current socioeconomic development of Bangladesh that can reflect the pattern succession of land use change following the previous development trend and explore the future land use structure.
The previously identified factors of land use changes acknowledge that these affect the land use changes of Bangladesh in 2010. According to the linear interpolation within the two time periods, the land uses for each year from 2010 to 2020 and 2010 to 2030 are then calculated.

Ecological Protection Priority

In ecological protection priority scenario, a number of useful measures would be implemented to protect the environment, while, at the same time, the growth of both population and GDP would be maintained at a lower rate. Under this scenario, forest areas would be reforested and grasslands would be re-cultivated with plants to reduce environmental degradation and protect the environment. Furthermore, built-up area expansion would be maintained at a lower rate.

Economic Growth

Under the economic growth scenario, the national average economic growth in the form of industrial structure; technology innovation; and adoption of policies to advocate expansions of urbanization, population and economy would be given preference. Under this projection, the built-up area will increase at a relatively higher speed; both the population growth rate and GDP growth rate will be higher than they are in baseline scenario; and more cultivated land will be converted to built-up area over time.

Spatial Allocation and Transition Probability of Land Use Change

The spatial allocation of land use change is the distribution of location specific quantitative land use change according to weight coefficients of driving factors and the scenarios. The decision rule of structural land use change is required in the spatial allocation of land use change. The value of decision rule is between 0 and 1. Bigger values indicate it is more difficult for one certain land use type to change to another land use type, and vice versa (e.g., urban land that is unlikely be converted again back into agricultural land). It is important to use a time set that is different from the original period for future projection. Mas et al. [62] exhibits that, in IDRISI (CA_MARKOV and LCM), transition probability matrices are used to calculate the quantity of each land use for a desired date for land use projection. The transition probability matrix is calculated using a simple powering when the projected date is a multiple of the calibration period. For example, we used the calibration period of 10 years (2000 to 2010) so the base matrix is squared to change probabilities for 2020, and raised to the power of three for 2030.
The transition probability is depicted in Table 3, which exhibits the transfer direction of land use types from 2000 to 2010. Out of six land use types, built-up area, water area and forest area are the most stable classes, with 0.83, 0.70 and 0.63 probabilities, respectively, while, the most dynamic classes are unused land, cultivated land and grassland with transition probabilities of 0.24, 0.29, and 0.64, respectively. In these classes, the cultivated lands were mainly transformed into built-up area and grassland also converted into built-up area.

2.3.2. DLS Model

For simulating the land use changes of Bangladesh in our scenarios, we selected and used the dynamics of land systems (DLS) model developed by Deng et al., 2008 [32]. The DLS model incorporates two special features: scenario analysis for future land demand (primarily setting up parameters value) and a spatial econometric regression between the explained and explanatory variables. The DLS considers that land use changes occur under land demand directed by socioeconomic developments in a region. Land use changes as pixel-specific changes are dominated by various factors according to Figure 2 (e.g., geophysical, socioeconomic, proximity or climatic factors). The results were utilized for simulating land use changes at any region. By executing the scenario analysis based on the hypothesis for key explanatory variables, the DLS model can improve the robustness of the simulation results [37]. During the present study, land use data from 2000 and 2010 were used to simulate the land use change of 2020 and 2030.
There exist three main modules in the DLS model. Firstly, a spatial regression module that analyzes the driving forces of land distribution for identifying the relationships between different land uses and influencing factors. Secondly, the scenario analysis module, which provides the changes in demand for different land use types at regional level during a given period. Thirdly, a spatial explicit allocation module, which is based mainly on a spatial regression analysis that can allocate land use changes from a regional level to the disaggregated grid cells [32]. Accordingly, for the current study, three processes occur in the simulation of spatial–temporal pattern changes of different land use types of Bangladesh: spatial regression analysis, analysis of the demand for different land use types and spatial allocation of different land use type changes. The spatial regression analysis, or the driving factor analysis, determines the statistical relationship between different land uses and their influencing/driving factors, which ultimately decides the weight coefficient matrix of driving factors of different land use changes [56]. Moreover, the driving factors may change over time due to the effect of conversion rules of original land use types as well as the changes in demand. For this study, we selected 11 driving factors for conducting the logistic regression analysis (Equation (1)) to obtain the relationships between the driving factors and the frequency of each land use type (results are shown in Table 4). The Logistic regression was utilized for estimating the probabilities of the existence of certain land use types in the cells. According to the regression results, the 11 driving factors can logically explain the spatial pattern of the six land use types. The relationship between the different factors affecting land use changes and their interaction between adjacent pixels with land uses can be incorporated by collinear diagnostic in the spatial lag terms (Equation (2)).
π = exp   ( b 0 +   j = 1 n b j X j + j = 1 n ρ W m n X j ) 1 + exp ( b 0 + j = 1 n b j X j + j = 1 n ρ W m n X j )
where, X is the vector of land use influencing factors, and π identifies the existence probability of a grid cell for the considered land use type j. Wmn is the spatial weight identifying the neighborhood between m and n. ρ is the estimated coefficient of spatial lag term of Xj. The increment of one unit of influencing factors is related to the increase of a unit increase of the influencing factors is associated with an increase in the exp(bj) plus exp(Wmn) of the probability of occurrence probability of the respective land use type. bj is the estimated coefficient of Xj, the residual constant of the equation.
The demand analysis or the scenario design furnishes one of the essential inputs in the DLS model. It determines the land use demand and calculates the dynamic demand changes of different land use types. The spatial allocation of land use change is an indispensable input parameter of the DLS model. The spatial allocation aims of determining the conversions that are allowed for a specific land use by a number of grid cells. It explains the temporal behavior of land use types or the status of grid cells [32]. For the current study, the spatial allocation and transition probability of land use change are shown in Table 3. For the simulating purpose, the whole land use structure was applied as input data and the development restricted areas were used as the restricted region.

2.3.3. Validation of DLS Model

Different land use models have been extensively utilized to judge possible land use dynamics and help future land management and policy. For validating the results of these models, it is common to check a comparison of actual and simulated land use pattern. We utilized Kappa coefficient to validate the DLS model. Kappa coefficient of the agreement is widely practiced method [63,64] for agreement assessment. Although we acknowledge that the Kappa coefficient has some bottlenecks, such as, Kappa value can be affected by skewed distributions of categories and the bias problem [65], and the kappa value for three or more categories is considered un-interpretable [66]. Despite the difficulties, it is the most extensively used coefficient for summarizing inter-rater agreement [67,68,69,70,71].
The land use datasets are categorical and we utilized Kappa for validating the accuracy of the DLS model [72]. For the current study, the land use data from 2000 and other factors data were employed as the base data for simulating the 2010 land use data. We then applied the Kappa to assess the agreement between actual and simulated land use change of 2010 through the following expression:
Kappa = (P0Pc)/(PpPc)
where, P0 is the percent correct for the output, Pc is the expected percent correct as by chance and Pp is the percent correct for perfect classification.

3. Results and Discussion

3.1. Performance of the DLS Model

The Kappa was employed to assess the agreement between the actual and simulated land use patterns in 2010 (Figure 3).
The value of Kappa statistic is 0.667. Based on the Kappa coefficient, Kappa value above 0.6 is considered a substantial [73,74,75] and/or good [76,77,78] assessment between actual and simulated results. Thereby, it shows that the applied DLS model in this study was appropriate to simulate the land use pattern of Bangladesh. Figure 3 shows the simulated and actual map of land use in 2010 are similar but still exhibit certain biases. Cultivated land, grassland, water area and unused land are similar to the actual map of 2010, while forest area and built-up area are poorly simulated and under-predicted by the DLS model.

3.2. Analysis of the Driving Factors for the Dynamics of Land Use Changes

The relationship of land use and different factors affecting changes of land is explored based on spatial regression analyses [32,37,40,41,45]. Table 4 summarizes the regression coefficients of the effect of influencing factors on different land use changes such as cultivated land, forest area, grassland, water area, built-up area and the unused land of Bangladesh in 2010. The standard errors inside the parentheses indicate that the influencing factors fluctuate on their effect on the dynamics of land use. Smaller standard error simply stronger factor effects on the land use dynamics. For example, population density and GDP affect strongly the change of cultivated land, water area, and built-up area. On the other hand, average rainfall strongly affects the grassland change.
The geophysical variables are important factors affecting LUCC in Bangladesh. The estimation illustrates that soil pH and DEM exerts an obvious influence on the LUCC, which means that higher soil pH and elevation lead to more grassland and forest area change. The climatic variables have little influence on LUCC in Bangladesh. Only average rainfall is found to have an influence on the LUCC. The variables measuring the effects of distances to divisional headquarters, the nearest highways, and the nearest secondary roads cannot be ignored. These variables have an effect on the cultivated land and forest area change. Socioeconomic variables are one of the major factors affecting LUCC in Bangladesh. Population density, urban population density, and GDP have influence on different LUCC. In highly populated areas, damage of cultivated land, forest, grassland and water areas is more dramatic. Meanwhile built-up area increases more in more populous areas. The estimation result under urban population density and GDP indicate the same results, such as regarding population density.

3.3. Dynamic Simulation of Land Changes

Cultivated land, forest area and built-up area are the major land use types in Bangladesh. Under the three scenarios, the changing trends of land uses in Bangladesh are shown in Figure 4 and Table 5. We observed that competition exists among the six land use types in different scenarios (baseline, ecological protection priority and economic growth scenarios). It mainly shows a decrease of cultivated land and increase of built-up areas. Meanwhile, when the ecological condition is considered, forest area and grassland area show a positive relationship with ecological protection. However, when economic growth is considered, a dramatic increase of built-up area is seen. These changes might be due to the effect of different influencing factors on different land use pattern (Table 5 and Figure 5).
Under the baseline scenario, every land use type would primarily expand or decrease according to their previous existing areas. By 2020, both forest area (13.35%) and grassland (12.89%) would be expanded from the other lands, i.e., water areas, unused lands, and cultivated lands. The main expanded grass land areas are located mainly in the northeast (Sylhet Division), and in forestry areas in the Chittagong hill tracts area. During baseline period, water areas are reduced (35%) as the water areas are affected by the annual change of rainfall, increased temperature, and population growth. By 2030, under baseline scenario, both forest area (13%) and built-up area (5%) would be increased, mainly due to the shrinkages of grassland (14%), and water area (34%). Under the baseline scenario, an increase of built-up area and decline of cultivated land is common feature due to the pressure of huge population. The dramatic change in land use types, including the reduction of water area by 35% and 34% in 2020 and 2030, respectively, will lead to shortage of water resources in Bangladesh.
Under the ecological protection priority scenario, the study assumes a land use that is under strict environmental protection policy from 2010 to 2030. Under the ecological protection priority scenario in 2020, both forest area (18%) and grassland (12%) would be expanded. The shrinkage of water areas (35%), cultivated land (2%) and unused area (45%) would be mostly responsible for this expansion. By 2030, under ecological protection priority scenario, forest area and grassland would be increased by 35% and 30%, respectively. On an average, water areas and cultivated land decrease by 6% and 35%, respectively, which indicates that the regulation has significant effect on checking the huge expansion of built-up land. Under the economic growth scenario, the expansion of built-up area is strikingly clear when compared to the other scenarios for 2020 and 2030, and with the actual map for 2010. This expansion of built-up area by 46% (2020) and 51% (2030) would be due to the shrinkage of cultivated area (8.24% and 7.78%, respectively), and the conversion of forest area (3.53% and 6.48%), and grassland (11.06% and 24.33%). In total, unused area is very small in Bangladesh but it was also converted to built-up area.
Figure 5 exhibits the land use pattern of Bangladesh under three scenarios for 2010–2030. The simulation results showed that LUCC during 2010–2030 under ecological protection priority scenario was driven by the substantial expansion of forest area and grassland areas and the decrease of cultivated land, water areas, and unused lands, which would have a significant impact on the environmental protection. This expansion trend for forest area and grassland occurs mainly in the northeast (Sylhet region), the Chittagong hill tracts and the southwest (the Sundarbans). Similarly, the results under the economic growth scenario indicated a steep increase in built-up areas with the loss of cultivated land, grassland, water areas and forest areas. Although overall in the whole country, forest areas and grassland would generally decrease, an increase would be observed in the northeast (Sylhet division). The forest areas in Sylhet region have been found to be unutilized for a long time. However, the government also plans to plant 5000 ha of forests in the Sylhet region according to the 7th Five Year Plan [60]. Social forestry programs in Bangladesh are beginning to become a social movement. Currently, the government is strongly considering extending this social forestry program, and an estimated 20,000 km of strip plantation might be planted. The country also plans to deploy the local government bodies and NGOs to coordinate afforestation programs at the grass root level.

4. Discussion

Compared with the baseline scenario, land use changes under ecological protection priority scenario would be marked by the expansion of forest area and grassland area. A noticeable amount of forest area and grassland areas would present in the northeast, southeast and southwest of Bangladesh in 2020 and 2030. According to the Prospective Plan of Bangladesh: Vision 2021 [58], Bangladesh Government wants to raise the forest area and grassland cover to 15% from the current 13%. Within this time frame, the government set a target of upgrading the tree coverage of 2.84 million hectares in forest area and grassland areas with the diversification of tree species for ecological balance as well as increasing forestry employment under the social and agro-forestry program. Currently, forest area and grassland are gaining importance in government policy h yielding more budgetary allocation for accomplishing forestry programs. For supporting forestry sector in climate change perspective, the government opens new programs, such as “Bangladesh Climate Change Trust Fund” and “Bangladesh Climate Change Resilient Fund”. The government also plans to conserve and protect forest and grassland ecosystems for environmental stability, improve biodiversity and increasing employment opportunity and alleviating poverty for forest-dependent people through people oriented afforestation programs. Moreover, the government plans to improve quality and density of existing forests and older plants through “enrichment planting” and “assisted natural regeneration”. According to the 7th Five-Year Plan (2016–2020), Bangladesh government plans to plant and/or replant an estimated 50,000 ha of hill forest, 5000 ha of plain forest and 30,000 ha of coastal area forest [60]. All these activities are helpful to increase the forest and grassland areas in the country. The findings from Hasan et al. stated that forest area declined 0.129% during 1976–2000 but increased 0.081% during 2000–2010 in Bangladesh [79]. The expansion of forest area and grassland area is found in the northeast, Chittagong hill tracts area and the southwest, while the grassland cover increased 0.009% (1976–2000) and then decreased 0.032% (2000–2010).
Comparing the baseline scenario with the economic growth scenario, the land cover change would be characterized by the dramatic expansion of built-up areas. Findings of the Prospective Plan of Bangladesh: Vision 2021 [58] suggests that Bangladesh wants to increase its GDP growth rate from 6.1% in 2010 to 10% by 2021. Similarly, the government wants to raise the GDP growth rate to 12% by 2030. According to the prospective plan, the industrial and manufacturing sectors will increase by double digits and account for about 37% of GDP by 2021. To become a middle-income country by 2021, industrial expansion must be accompanied by the agricultural sector. Hence, increasing pace of built-up area and industrialization will be an important feature in the future to address the incremental poor capacity of agricultural sector to absorb increasing labor force, cater the growing local demand for industrial goods and also taking advantage of emerging opportunities of the global market. Meanwhile, under economic growth scenario, expansion of built-up areas would be profound in the southeast of Bangladesh and the Chittagong hill tracts area, as the population of these areas is increasing. In 1972, the government announced about 50,000 ha forest area as reserve forest and 4000 ha was leased for rubber plantation [80]. This decision enticed migrating low land people into the Chittagong hill tracts. Moreover, it was a common perception of the government that the Chittagong hill tracts is a thinly populated area, which ultimately accelerated the government to emigrate lowland people into the Chittagong hill tracts [81] resulting in an estimated 200,000–450,000 people being resettled in the Chittagong hill tracts, increasing population density and population pressure [82].
The built-up area increase is also associated with increasing housing demand in urban areas that can stimulate the land use pattern change. On the contrary, if the high population density in cities of Bangladesh cannot be accommodated, then sprawl might occur. This rapid spatial expansion of cities will exert huge pressure on cultivated land conversion to built-up areas in the future. On the other hand, rapid and unplanned urban area development and management is a challenge for the government. The government set a target of maintaining balance distribution of urban centers throughout the country with the cleaner urban environment. It is estimated that, by 2021, nearly one-third of the total population in Bangladesh will live in urban areas. Moreover, the urbanization in Bangladesh is characterized by the predominance of few urban centers and uneven distribution of urban population. A large proportion of urban population is concentrated in Dhaka, Chittagong and Khulna Metropolitan areas, which comprise about 54% of the total urban population [58]. Seizing of land by the powerful, changing ownership through false statements by the corrupt officials, and the subdivision of lands into un-economical parcels by inheritance make land in Bangladesh a source of constant violence [83]. Moreover, national politicians and bureaucrats in Bangladesh have large land ownership and often rely on rich landholder farmers to gain political support [84]. Bangladesh government has changed its policy to redistribute fallow land to landless households since 1987, although the system is still controlled by local politicians and large landowners [85].
As cultivated land loss amounts to 1% annually [58], to ensure food security for the huge population, sustainable land management is essential. According to the Prospective Plan of Bangladesh (2010–2021), the government formulated a policy to promote sustainable urban land management, which should be environmentally sound ([58], p.76). Moreover, government plans to take effective measures for protecting Bangladesh from adverse climatic effects and global warming. At the same time, it targets protecting the environment from manmade threats such as deforestation, air pollution, over fishing, river encroachment, unplanned building construction, conversion of water areas into agricultural land and other purposes and population pressure on the land. Additionally, the government of Bangladesh also identified major policy thrust areas as the maximum utilization of available lands, capturing and reversing the degraded lands. In this context, to prepare an integrated action plan, simulation results of the current study of Bangladesh can provide decision making references spatial land management planning and control practices.
Due to high population growth rate, increasing income growth rate and increasing per capita income are important factors affecting intensive fixed land area competition in Bangladesh. The land use patterns are changing in complex and different ways in Bangladesh. Thus, meaningful land use policy is necessary to ensure the best possible use of land and water bodies, restrict their misuse, protect the gradual decreasing trend of agricultural land for feeding the growing population, ensure best utilization of fallow lands, and control unsubstantiated land business by the landowners [86]. As a result of a huge population density of the country, land is the scarcest production factor in Bangladesh. This situation enhances the skyrocketing prices of land all over the country, especially in metropolitan cities. Thus, Bangladesh government is strongly considering this when developing future growth strategies and trying to manage the land for the sustainable rapid growth of the country. Thus, sound land management is thought to have a direct effect on poverty reduction and social welfare [59]. For controlling the high conversion of agricultural land into nonagricultural purposes, determining the best land use, and enhancing the environmental sustainability of land use, the government formulated and adopted a National Land Use Policy in 2001 [59]. Moreover, the government is trying to develop a database of all land resources, and land-zoning information in some selected areas and other environmentally endangered area. Bangladesh has a great shortage of vacant land, which pushes land price even higher, developing a nontransparent land market throughout the country. This makes it impossible for most urban residents to buy land and build their own houses [60]. Thus, the current research findings will help planners and decision makers formulate future sustainable land use policy in Bangladesh.
Bangladesh formally joined China’s “one belt, one road” initiative in October 2016 [87]. The “one belt, one road” initiative aims to maintain connectivity in five aspects: infrastructure, resources development, financial and industrial cooperation, and people to people bond [88]. The land based “one belt one road” will connect 14 stops including the port city Chittagong, Bangladesh. Thus, a huge opportunity exists for Bangladesh in terms of the development of many aspects, and future land use pattern development is important for Bangladesh to strengthen the “one belt, one road” initiative of China. The current research findings can be utilized as a guideline to develop future land use policy.
It is recognized that no single model can be sufficient enough to describe the different process of land use change pattern [89]. Assessment of the validity of recent studies has identified that most spatial models still have high uncertainty [90]. We selected several variables such as geophysical, socioeconomic, climatic and proximity variables, but there are still some other factors that are difficult to quantify. Selection of factors and indicators, to a certain extent, may cause differences in the simulation results; for example, housing size changes in built-up areas may have an influence on spatial land use pattern in respect of housing demand, and policy related issues (such as land use zoning, infrastructural policy) impact on the land use pattern change. However, in this study, we do not deal with policy related variables. A future study could employ the current study variables together with policy and other variables in the simulation of LUCC in Bangladesh, yielding more fruitful decision-making recommendations.

5. Conclusions

It is of great importance to explore the dynamics and spatial pattern of land use and land cover changes in a land-hungry country such as Bangladesh. The present case study of Bangladesh recognized related driving factors, including geophysical, climatic, proximity and socioeconomic variables, which were exploited using a dynamics of land systems (DLS) model for simulating land use demand over the next 20 years (2010–2030).
The scenario simulation offers a distinct picture of land use and land cover changes. The results suggest that, following the normal trend of land use change according to the previous years and areas under the baseline scenario, forest area would be augmented and water area as well as cultivated lands would be decreased. The water area shrinkage will have consequences for the water crisis in Bangladesh. Subsequently, when the ecological protection priority scenario is considered, cultivated lands and water areas convert to forest area and grassland, which will become the main feature of land use and land cover change in the future. Under the economic growth scenario, steep inflation of built-up areas will become the main feature of land use and land cover change in Bangladesh in future. Thus, to protect ecological conservation and cultivated lands from conversion, and strictly control the expansion of built-up areas, important land use adjustments are necessary to accelerate overall ecological and sustainable economic development. Thus, the output of the current research will generate valuable information for supporting the systemic management of the existing land resources. At the same time, the findings will also assist in planning and policy development as a future guide of different land conversion directions targeting the quantities of each land use types. The results will also help to present the tradeoffs among different land use patterns in specific regions of Bangladesh.

Acknowledgments

The dataset of this research was supported by the National Natural Science Funds of China for Distinguished Young Scholars (Grant No. 71225005), and it was financially supported by the STS-Network Project of Chinese Academy of Sciences (Grant No. KFJ-EW-STS-058).

Author Contributions

In this article, Shaikh Shamim Hasan contributed to research design and organized the research flow, data preparation, analysis and interpretation; Xiangzheng Deng contributed to the result analysis and interpretation; Zhihui Li contributed to the interpretation of results; and Dongdong Chen contributed to the DLS simulation. All the authors contributed to the writing of the article.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Hasan, S.S.; Hossain, M.; Sultana, S.; Ghosh, M.K. Women’s involvement in income generating activities and their opinion about its contribution: A study of Gazipur District, Bangladesh. Sci. Innov. 2015, 3, 72–80. [Google Scholar] [CrossRef]
  2. Sultana, S.; Hasan, S.S. Impact of Micro-Credit on Economic Empowerment of Rural Women. Agriculturists 2010, 8, 43–49. [Google Scholar] [CrossRef]
  3. Planning Commission of Bangladesh. Population and Housing Census, Preliminary Result of 2011; Bangladesh Bureau of Statistics (BBS): Dhaka, Bangladesh, 2014. Available online: http://203.112.218.65/WebTestApplication/userfiles/Image/National%20Reports/SED_REPORT_Vol-4.pdf (accessed on 30 December 2012).
  4. Uddin, K.; Gurung, D.R. Land cover change in Bangladesh—A knowledge based classification approach. In Proceedings of the 10th International Symposium on High Mountain Remote Sensing Cartography, ICIMOD, Kathmandu, Nepal, 8–11 September 2010. [Google Scholar]
  5. Biro, K.; Pradhan, B.; Buchroithner, M.; Makeschin, F. Land use/land cover change analysis and its impact on soil properties in the Northern part of Gadarif region, Sudan. Land Degrad. Dev. 2013, 24, 90–102. [Google Scholar] [CrossRef]
  6. Leh, M.; Bajwa, S.; Chaubey, I. Impact of land use change on erosion risk: An integrated remote sensing, geographic information system modeling methodology. Land Degrad. Dev. 2013, 24, 409–421. [Google Scholar] [CrossRef]
  7. Wu, X.; Shen, Z.; Liu, R.; Ding, Z. Land Use/Cover Dynamics in Response to Changes in Environmental and Socio-Political Forces in the Upper Reaches of the Yangtze River, China. Sensors 2008, 8, 8104–8122. [Google Scholar] [CrossRef] [PubMed]
  8. Barua, S.K.; Haque, S.M.S. Soil characteristics and carbon sequestration potentials of vegetation in degraded hills of Chittagong, Bangladesh. Land Degrad. Dev. 2013, 24, 63–71. [Google Scholar] [CrossRef]
  9. Zhang, J.J.; Fu, M.C.; Zeng, H.; Geng, Y.H.; Hassani, F.P. Variations in ecosystem service values and local economy in response to land use: A case study of Wuhan, China. Land Degrad. Dev. 2013, 24, 236–249. [Google Scholar] [CrossRef]
  10. Steffen, W.; Crutzen, P.J.; McNeill, J.R. The Anthropocene: Are Humans Now Overwhelming the Great Forces of Nature. AMBIO J. Hum. Environ. 2007, 36, 614–621. [Google Scholar] [CrossRef]
  11. Mendoza, M.E.; Granados, E.L.; Geneletti, D.; Pérez-Salicrup, D.R.; Salinas, V. Analysing land cover and land use change processes at watershed level: A multi temporal study in the Lake Cuitzeo Watershed, Mexico (1975–2003). Appl. Geogr. 2011, 31, 37–50. [Google Scholar] [CrossRef]
  12. Arsanjani, J.J.; Helbich, M.; Kainz, W.; Boloorani, A.D. Integration of logistic regression, Markov chain and cellular automata models to simulate urban expansion. Int. J. Appl. Earth Obs. Geoinform. 2013, 21, 265–275. [Google Scholar] [CrossRef]
  13. Ellis, E. Land Use and Land Cover Change. 2013. Available online: http://www.eoearth.org/view/article/154143 (accessed on 20 March 2013).
  14. Shapla, T.; Park, J.; Hongo, C.; Kuze, H. Agricultural Land Cover Change in Gazipur, Bangladesh, in Relation to Local Economy Studied Using Landsat Images. Adv. Remote Sens. 2015, 4, 214–223. [Google Scholar] [CrossRef]
  15. Kumar, M.; Ghosh, M.K. Land cover change detection of Hatiya Island, Bangladesh, using remote sensing techniques. J. Appl. Remote Sens. 2012, 6, 063608. [Google Scholar] [CrossRef]
  16. Zaman, S.; Katoh, M. Assessment of forest cover change in tropical moist deciduous forest in Takurgaon, Bangladesh using ALOS data. J. For. Plan. 2011, 16, 285–292. [Google Scholar]
  17. Rahman, M.M.; Begum, S. Land cover change analysis around the Sundarbans Mangrove forest of Bangladesh using remote sensing and GIS application. J. Sci. Found. 2011, 9, 95–107. [Google Scholar] [CrossRef]
  18. Islam, R.M.; Hasan, M.Z. Land use changing pattern and challenges for agricultural land: A study on Rajshahi district. J. Life Earth Sci. 2011, 6, 69–74. [Google Scholar] [CrossRef]
  19. Mamun, A.A.; Mahmood, A.; Rahman, M. Identification and Monitoring the Change of Land Use Pattern Using Remote Sensing and GIS: A Case Study of Dhaka City. IOSR J. Mech. Civ. Eng. 2013, 6, 20–28. [Google Scholar] [CrossRef]
  20. Roy, S.; Farzana, K.; Papia, M.; Hasan, M. Monitoring and Prediction of Land Use/Land Cover Change using the Integration of Markov Chain Model and Cellular Automation in the Southeastern Tertiary Hilly Area of Bangladesh. Int. J. Sci. Basic Appl. Res. 2015, 24, 125–148. [Google Scholar]
  21. Ahmed, S.J.; Bramley, G.; Verburg, P.H. Key Driving Factors Influencing Urban Growth: Spatial-Statistical Modelling with CLUE-s. In Dhaka Megacity: Geospatial Perspectives on Urbanization, Environment and Health; Dewan, A., Corner, R., Eds.; Springer Science & Business Media: New York, NY, USA, 2014. [Google Scholar]
  22. Islam, M.S.; Ahmed, R. Land use change prediction in Dhaka city using GIS aided Markov Chain Modeling. J. Life Earth Sci. 2011, 6, 81–89. [Google Scholar] [CrossRef]
  23. Corner, R.J.; Dewan, A.M.; Chakma, S. Monitoring and Prediction of Land-Use and Land-Cover (LULC) Change. In Dhaka Megacity: Geospatial Perspectives on Urbanisation, Environment and Health; Dewan, A., Corner, R., Eds.; Springer Science & Business Media: New York, NY, USA, 2014. [Google Scholar]
  24. The World Bank (WB). Sustainable Land Management. Challenges, Opportunities, and Trade-Offs; International Bank for Reconstruction and Development (IBRD): Washington, DC, USA, 2006. [Google Scholar]
  25. Wang, X.; Zhao, X.L.; Zhang, Z.X.; Yi, L.; Zuo, L.J.; Wen, Q.K.; Liu, F.; Xu, J.Y.; Hu, S.G.; Liu, B. Assessment of soil erosion change and its relationships with land use/cover change in China from the end of the 1980s to 2010. Catena 2016, 137, 256–268. [Google Scholar] [CrossRef]
  26. Xie, H.L.; He, Y.; Xie, X. Exploring the factors influencing ecological land change for China’s Beijing-Tianjin-Hebei Region using big data. J. Clean. Prod. 2017, 142, 677–687. [Google Scholar] [CrossRef]
  27. Xie, H.L.; Yao, G.R.; Liu, G.Y. Spatial evaluation of ecological importance based on GIS for environmental management: A case study in Xingguo County of China. Ecol. Indic. 2015, 51, 3–12. [Google Scholar] [CrossRef]
  28. Yao, G.; Xie, H. Rural restructuring in ecologically fragile mountainous areas of southern China: A case study of Changgang Town, Jiangxi Province. J. Rural Stud. 2016, 47, 435–448. [Google Scholar] [CrossRef]
  29. Xie, H.L.; Liu, Z.F.; Wang, P.; Liu, G.Y.; Lu, F.C. Exploring the mechanisms of ecological land change based on the spatial autoregressive model: A case study of the Poyang Lake eco-economic zone, China. Int. J. Environ. Res. Public Health. 2014, 11, 583–599. [Google Scholar] [CrossRef] [PubMed]
  30. Xie, Z.L.; Liu, J.Y.; Ma, Z.W.; Duan, X.F.; Cui, Y.P. Effect of surrounding land-use change on the wetland landscape pattern of a natural protected area in Tianjin, China. Int. J. Sustain. Dev. World Ecol. 2012, 19, 16–24. [Google Scholar] [CrossRef]
  31. Cao, V.; Margin, M.; Favis, B.D.; Deschenes, L. Aggregated indicator to assess land use impacts in life cycle assessment (LCA) based on the economic value of ecosystem services. J. Clean. Prod. 2015, 94, 56–66. [Google Scholar] [CrossRef]
  32. Deng, X.Z.; Su, H.B.; Zhan, J.Y. Integration of multiple data sources to simulate the dynamics of land systems. Sensors 2008, 8, 620–634. [Google Scholar] [CrossRef] [PubMed]
  33. Manson, S.M. Challenges in evaluating models of geographic complexity. Environ. Plan. B. 2007, 34, 245–260. [Google Scholar] [CrossRef]
  34. Yu, Q.; Wu, W.; Verburg, P.H.; van Vliet, J.; Yang, P.; Zhou, Q.; Tang, H. A survey-based exploration of land-system dynamics in an agricultural region of Northeast China. Agric. Syst. 2013, 121, 106–116. [Google Scholar] [CrossRef]
  35. VanVilet, J.; Bregt, A.K.; Brown, D.G.; Delden, H.V.; Heckbert, S.; Verburg, P.H. A review of current calibration and validation practices in land-change modeling. Environ. Model. Softw. 2016, 82, 174–182. [Google Scholar] [CrossRef]
  36. Sterk, B.; van Ittersum, M.K.; Leeuwis, C. How, when, and for what reasons does land use modelling contribute to societal problem solving? Environ. Model. Softw. 2011, 26, 310–316. [Google Scholar] [CrossRef]
  37. Deng, X.; Jiang, Q.; Zhan, J.; He, S.; Lin, Y. Simulation on the dynamics of forest area changes in northeast China. J. Geogr. Sci. 2010, 20, 495–509. [Google Scholar] [CrossRef]
  38. Jiang, Q.; Cheng, Y.; Jin, Q.; Deng, X.; Qi, Y. Simulation of Forestland Dynamics in a Typical Deforestation and Afforestation Area under Climate Scenarios. Energies 2015, 8, 10558–10583. [Google Scholar] [CrossRef]
  39. Jiang, Q.; Ma, E.; Zhan, J.; Shi, N. Seasonal and inter annual variation in energy balance in the semi-arid grassland area of China. Adv. Meteorol. 2015, 2015, 120620. [Google Scholar] [CrossRef]
  40. Li, Z.; Deng, X.; Wu, F.; Hasan, S.S. Scenario Analysis for Water Resources in Response to Land Use Change in the Middle and Upper Reaches of the Heihe River Basin. Sustainability 2015, 7, 3086–3108. [Google Scholar] [CrossRef]
  41. Zhen, L.; Deng, X.; Wei, Y.; Jiang, Q.; Lin, Y.; Helming, K.; Wang, C.; König, H.; Hu, J. Future land use and food security scenarios for the Guyuan district of remote western China. iForest 2014, 7, 362. [Google Scholar] [CrossRef]
  42. Bangladesh Bureau of Statistics (BBS). Statistical Pocket Book of Bangladesh 2014; Planning Division, Ministry of Planning, Government of the People’s Republic of Bangladesh, 2015. Available online: http://203.112.218.65/WebTestApplication/userfiles/Image/LatestReports/PB2014.pdf (accessed on 15 April 2016).
  43. Globeland30. Available online: www.globallandcover.com (accessed on 25 March 2017).
  44. Ran, Y.H.; Li, X. First China—Comments on global land cover map at 30-m resolution. Sci. China Earth Sci. 2015, 58, 1677–1678. [Google Scholar] [CrossRef]
  45. Yan, H.; Zhan, J.; Jiang, Q. Scenario simulation of changes of forest land in Poyang Lake watershed. Procedia Environ. Sci. 2010, 2, 1469–1478. [Google Scholar]
  46. Han, H.; Yang, C.; Song, J. Scenario Simulation and the Prediction of Land Use and Land Cover Change in Beijing, China. Sustainability, 2015, 7, 4260–4279. [Google Scholar] [CrossRef]
  47. Liu, J.Y.; Liu, M.L.; Tian, H.Q.; Zhuang, D.F.; Zhang, Z.X.; Zhang, W.; Tang, X.M.; Deng, X.Z. Spatial and temporal patterns of China’s cropland during 1990–2000: An analysis based on Landsat TM data. Remote Sens. Environ. 2005, 98, 442–456. [Google Scholar] [CrossRef]
  48. Turner, B.L.; Moss, R.H.; Skole, D. Relating Land Use and Global Land-Cover Change: A Proposal for an IGBP-HDP Core Project; International Geosphere-Biosphere Programme: Stockholm, Sweden, 1993; Available online: http://www.ciesin.org/ (accessed on 8 January 2005).
  49. Yue, T.X.; Wang, Y.A.; Liu, J.Y.; Chen, S.P.; Qiu, D.S.; Deng, X.Z.; Liu, M.L.; Tian, Y.Z.; Su, B.P. Surface modeling of human population distribution in China. Ecol. Model. 2005, 181, 461–478. [Google Scholar] [CrossRef]
  50. Doll, C.N.H.; Muller, J.P.; Morley, J.G. Mapping regional economic activity from night-time light satellite imagery. Ecol. Econ. 2006, 57, 75–92. [Google Scholar] [CrossRef]
  51. Eric, P.J.; Beurs, K.M.B.; Hartkamp, A.D. Kriging and thin plate splines for mapping climate variables. Int. J. Appl. Earth Obs. Geoinform. 2001, 3, 146–154. [Google Scholar]
  52. Batjes, N.H.; Ribeiro, E.; van Oostrum, A.; Leenaars, J.; Hengl, T.; de Mendes, J.J.; Wo, S.I.S. Serving standardised soil profile data for the world. Earth Syst. Sci. Data Discuss. 2016. [Google Scholar] [CrossRef]
  53. Ribeiro, E.; Batjes, N.H.; Leenaars, J.G.B.; Van Oostrum, A.J.M.; de Mendes, J.J. Towards the Standardization and Harmonization of World Soil Data: Procedures Manual ISRIC World Soil Information Service (WoSIS Version 2.0); ISRIC—World Soil Information: Wageningen, The Netherlands, 2015; p. 110. Available online: http://www.isric.org/sites/default/files/isric_report_2015_03.pdf (accessed on 1 April 2015).
  54. Bangladesh Bureau of Statistics (BBS). Statistical Year Book of Bangladesh 2011; Planning Division, Ministry of Planning, Government of the People’s Republic of Bangladesh, 2012. Available online: http://203.112.218.65/WebTestApplication/userfiles/Image/LatestReports/YB2011.pdf (accessed on 15 September 2012).
  55. Stevens, F.R.; Gaughan, A.E.; Linard, C.; Tatem, A.J. Disaggregating Census Data for Population Mapping Using Random Forests with Remotely-Sensed and Ancillary Data. PLoS ONE 2015, 10, e0107042. [Google Scholar] [CrossRef] [PubMed]
  56. Bai, W.Q.; Zhang, Y.M.; Yan, J.Z.; Zhang, Y.L. Simulation of land use dynamics in the upper reaches of the Daduriver. Geogr. Res. 2005, 24, 206–212. [Google Scholar]
  57. Pontius, R.G.; Peethambaram, S.; Castella, J.C. Comparison of three maps at multiple resolutions: A case study of land change simulation in Cho Don District, Vietnam. Ann. Assoc. Am. Geogr. 2011, 101, 45–62. [Google Scholar] [CrossRef]
  58. Planning Commission of Bangladesh. Prospective Plan of Bangladesh 2010–2021: Making Vision 2021 a Reality; General Economics Division, Planning Commission, Ministry of Planning: Dhaka, Bangladesh, 2012. Available online: http://www.plancomm.gov.bd/wp-content/uploads/2013/09/Perspective-Plan-of-Bangladesh.pdf (accessed on 25 September 2013).
  59. Planning Commission of Bangladesh. The 6th Five Year Plan (FY2011-FY2015); Ministry of Planning: Dhaka, Bangladesh, 2015. Available online: http://www.plancomm.gov.bd/sixth-five-year-plan/ (accessed on 26 September 2011).
  60. Planning Commission of Bangladesh. The 7th Five Year Plan: Background Study; Ministry of Planning: Dhaka, Bangladesh, 2016. Available online: http://www.plancomm.gov.bd/7th-five-year-plan/ (accessed on 23 February 2015).
  61. The World Bank Data Portal. 2016. Available online: http://data.worldbank.org/country/bangladesh?view=chart (accessed on 27 July 2016).
  62. Mas, J.F.; Kolb, M.; Paegelow, M.; Olmedo, M.T.C.; Houet, T. Inductive pattern-based land use/cover change models: A comparison of four software packages. Environ. Model. Softw. 2014, 51, 94–111. [Google Scholar] [CrossRef]
  63. Cohen, J.A. Coefficient of agreement for nominal scales. Educ. Psychol. Meas. 1960, 20, 37–46. [Google Scholar] [CrossRef]
  64. Congalton, R.G. A review of assessing the accuracy of classifications of remotely sensed data. Remote Sens. Environ. 1991, 37, 35–46. [Google Scholar] [CrossRef]
  65. Eugenio, D.B.; Glass, M. The Kappa Statistic: A Second Look. Comput. Linguist. 2004, 30, 95–101. [Google Scholar] [CrossRef]
  66. Kraemer, H.C.; Periyakoil, V.S.; Noda, A. Kappa coefficients in medical research. Stat. Med. 2002, 21, 2109–2129. [Google Scholar] [CrossRef] [PubMed]
  67. Hartling, L.; Bond, K.; Santaguida, P.L.; Viswanathan, M.; Dryden, D.M. Testing a tool for the classification of study designs in systematic reviews of interventions and exposures showed moderate reliability and low accuracy. J. Clin. Epidemiol. 2011, 64, 861–871. [Google Scholar] [CrossRef] [PubMed]
  68. Ingenhoven, T.J.; Duivenvoorden, H.J.; Brogtrop, J.; Lindenborn, A.; van den Brink, W.; Passchier, J. Interrater reliability for Kernberg’s structural interview for assessing personality organization. J. Personal. Disord. 2009, 23, 528–534. [Google Scholar] [CrossRef] [PubMed]
  69. Ansari, N.N.; Naghdi, S.; Forogh, B.; Hasson, S.; Atashband, M.; Lashgari, E. Development of the Persian version of the Modified Modified Ashworth Scale: Translation, adaptation, and examination of interrater and intrarater reliability in patients with post stroke elbow flexor spasticity. Disabil. Rehabil. 2012, 34, 1843–1847. [Google Scholar] [CrossRef] [PubMed]
  70. Hsu, L.M.; Field, R. Interrater agreement measures: Comments on kappa η, Cohen’s kappa, Scott’s π and Aickin’sα. Underst. Stat. 2003, 2, 205–219. [Google Scholar] [CrossRef]
  71. Zwick, R. Another look at interrater agreement. Psychol. Bull. 1988, 103, 374–378. [Google Scholar] [CrossRef] [PubMed]
  72. Hagen-Zanker, A.; Lajoie, G. Neutral models of landscape change as benchmarks in the assessment of model performance. Landsc. Urban Plan. 2008, 86, 284–296. [Google Scholar] [CrossRef]
  73. Saraux, A.; Tobón, G.J.; Benhamou, M.; Devauchelle-Pensec, V.; Dougados, M.; Mariette, X.; Berenbaum, F.; Chiocchia, G.; Rat, A.C.; Schaeverbeke, T. Potential classification criteria for rheumatoid arthritis after two years: Results from a French multi center cohort. Arthritis Care Res. 2013, 65, 1227–1234. [Google Scholar] [CrossRef] [PubMed]
  74. Anthony, J.; Viera, M.D.; Joanne, M.G. Understanding Inter observer Agreement: The Kappa Statistic. Fam. Med. 2005, 37, 360–363. [Google Scholar]
  75. Landis, J.R.; Koch, G.G. The measurement of observer agreement for categorical data. Biometrics 1977, 33, 159–174. [Google Scholar] [CrossRef] [PubMed]
  76. Altman, D.G. Practical Statistics for Medical Research; Chapman and Hall: London, UK, 1991. [Google Scholar]
  77. Cohen, J. Weighted kappa: Nominal scale agreement provision for scaled disagreement or partial credit. Psychol. Bull. 1968, 70, 213–220. [Google Scholar] [CrossRef] [PubMed]
  78. Fleiss, J.L.; Cohen, J. The equivalence of weighted kappa and the intraclass correlation coefficient as measures of reliability. Educ. Psychol. Meas. 1973, 33, 613–619. [Google Scholar] [CrossRef]
  79. Hasan, M.N.; Hossain, M.S.; Bari, M.A.; Islam, M.R. Agricultural Land Availability in Bangladesh; SRDI: Dhaka, Bangladesh, 2013. Available online: http://www.srdi.gov.bd/wp-content/uploads/2014/03/Agricultural-land-availability-in-Bangladesh-monograph-1.pdf (accessed on 11 May 2014).
  80. Mohsin, A. The Politics of Nationalism: The Case of Chittagong Hill Tracts, Bangladesh; The University Press Limited: Dhaka, Bangladesh, 1997. [Google Scholar]
  81. Rasul, G.; Thapa, G.B. State Policies, Praxies and Land-Use in the Chittagong Hill Tracts of Bangladesh; IIED Working Paper; International Institute for Environment and Development: London, UK, 2005; Available online: http://hdl.handle.net/10535/6189 (accessed on 25 August 2010).
  82. Roy, R.D. Sustainable and equitable resource management in the Chittagong Hill Tracts. In Farming Practices and Sustainable Development in the Chittagong Hill Tracts; Khan, N.A., Alam, M.K., Khisa, S.K., Millat-e-Mustafa, M., Eds.; CHTDB and VFFP-IC: Chittagong, Bangladesh, 2002; pp. 135–154. [Google Scholar]
  83. Seabrook, J. In the city of hunger: Barisal Bangladesh. Race Class. 2013, 51, 39–58. [Google Scholar] [CrossRef]
  84. Hossain, M.; Jones, S. Production, Poverty and the Co-operative Ideal: Contradictions in Bangladesh Rural Development Policy. In Rural Development and the State: Contradictions and Dilemmas in Developing Countries; Lea, D., Chaudhuri, D., Eds.; Methuen: London, UK, 1983; pp. 161–185. [Google Scholar]
  85. Devine, J. NGOs, politics and grassroots mobilisation: Evidence from Bangladesh. J. South Asian Dev. 2006, 1, 77–99. [Google Scholar] [CrossRef]
  86. Barakat, A.; Ara, R.; Taheruddin, M.; Hoque, S.; Islam, N. Towards a Feasible Land Use Policy of Bangladesh; Human Development Research Centre (HDRC): Dhaka, Bangladesh, 2007. [Google Scholar]
  87. Bangladesh Formally Joins China’s Flagship ‘One Belt One Road’ Initiative. Available online: http://bdnews24.com/bangladesh/2016/10/15/bangladesh-formally-joins-chinas-flagship-one-belt-one-road-initiative (accessed on 25 October 2016).
  88. Huang, Y. Understanding China’s Belt & Road Initiative: Motivation, framework and assessment. China Econ. Rev. 2016, 40, 314–321. [Google Scholar]
  89. Verburg, P.H.; Overmars, K.P. Dynamic simulation of land-use change trajectories with the CLUE-s model. Model. Land Use Chang. 2007, 90, 321–337. [Google Scholar]
  90. Pontius, R.G., Jr.; Huffaker, D.; Denman, K. Useful techniques of validation for spatially explicit land-change models. Ecol. Model. 2004, 179, 445–461. [Google Scholar] [CrossRef]
Figure 1. Boundary map and surrounding areas of Bangladesh.
Figure 1. Boundary map and surrounding areas of Bangladesh.
Sustainability 09 00505 g001
Figure 2. Frame work of the Dynamics of Land System (DLS) Model.
Figure 2. Frame work of the Dynamics of Land System (DLS) Model.
Sustainability 09 00505 g002
Figure 3. Comparison of: actual (a); and simulated (b) land use situation in Bangladesh, 2010.
Figure 3. Comparison of: actual (a); and simulated (b) land use situation in Bangladesh, 2010.
Sustainability 09 00505 g003
Figure 4. Land use structure changes under baseline, ecological protection and economic growth scenarios.
Figure 4. Land use structure changes under baseline, ecological protection and economic growth scenarios.
Sustainability 09 00505 g004
Figure 5. Projected land use situation under: (a,a’) the baseline scenario; (b,b’) the ecological protection priority scenario; and (c,c’) economic growth scenario of Bangladesh.
Figure 5. Projected land use situation under: (a,a’) the baseline scenario; (b,b’) the ecological protection priority scenario; and (c,c’) economic growth scenario of Bangladesh.
Sustainability 09 00505 g005
Table 1. Reclassification of land uses of Bangladesh.
Table 1. Reclassification of land uses of Bangladesh.
CodeLand useDescription
0Cultivated landOriginal data include both rice and non-irrigated uplands.
1Forest areaNatural or planted forests with canopy covers greater than 30%; land covered by trees less than 2 m high, with a canopy cover greater than 40%; land covered by trees with canopy cover between 10% and 30%; and land used for teagardens, orchards and nurseries.
2GrasslandLands covered by herbaceous plants with coverage greater than 5% and land mixed rangeland with the coverage of shrub canopies less than 10%.
3Water areaLand covered by natural water bodies or land with facilities for irrigation and water reservation, including rivers, canals, lakes, beaches and shorelines, and bottomland.
4Built-up areaLand used for urban and rural settlements, industry and transportation.
5Unused land All other lands.
Table 2. List of influencing factors for considered during land use change study in Bangladesh.
Table 2. List of influencing factors for considered during land use change study in Bangladesh.
Factor TypeInfluencing FactorDefinition
Geophysical variablesSoil pHpH value of the soil
ElevationDigital Elevation Model (m) (DEM)
Climatic variablesAir temperatureMean annual temperature (°C)
Hours of sunshineHours of sunshine rectified to account for the spatial variability of solar radiation (mean annual) (i.e., topographic effects)
Average rainfallMean annual rainfall (mm)
Proximity variablesDistance to divisional head quarterGeometric distance to the nearest divisional head quarter (km)
Distance to nearest highwaysDistance to the nearest highway (km)
Distance to nearest secondary highwaysDistance to the second nearest highway (km)
Socioeconomic variablesPopulation densityInterpolated population density (persons/km2) based on the population distribution model [32,49]
GDPInterpolated values of gross domestic product (GDP; million USD/km based on spatially explicit analyses of the relationship between economic growth and factors that might affect economic growth [50]
Urban population densityInterpolated urban population density (persons/km2) based on the population distribution model
Table 3. Transition probability of land use types in Bangladesh from 2000 to 2010.
Table 3. Transition probability of land use types in Bangladesh from 2000 to 2010.
Cultivated LandForest AreaGrasslandWater AreaBuilt-Up AreaUnused Land
Cultivated land0.970.020.040.150.120.04
Forest area0.020.930.080.020.010.00
Grassland0.000.020.820.030.000.35
Water area0.000.000.020.770.010.06
Built-up area0.000.010.010.020.850.00
Unused land0.000.000.010.000.000.54
Table 4. Regression coefficients of the influencing factors of land uses in Bangladesh. Standard errors are included in parentheses.
Table 4. Regression coefficients of the influencing factors of land uses in Bangladesh. Standard errors are included in parentheses.
Driving FactorsCultivated LandForest AreaGrasslandWater AreaBuilt-Up AreaUnused Land
Soil pH0.020 (0.001)−0.014 (0.001)0.000 (0.001)−0.015 (0.001)−0.001 (0.001)−0.006 (0.013)
DEM−0.015 (0.000)0.001 (0.000)−0.008 (0.001)−0.003 (0.001)−0.017 (0.001)−0.033 (0.047)
Average temperature−0.179 (0.018)−3.256 (0.053)2.718 (0.053)0.044 (0.045)−0.004 (0.027)3.194 (2.026)
Average sunshine hours−1.779 (0.096)14.676 (0.181)2.255 (0.261)−1.501 (0.257)−6.494 (0.173)4.305 (8.622)
Average rainfall−0.012 (0.000)0.004 (0.000)0.012 (0.000)−0.015 (0.001)−0.001 (0.000)0.063 (0.016)
Distance to divisional HQ0.003 (0.000)0.002 (0.000)−0.003 (0.001)−0.009 (0.001)−0.007 (0.000)−0.028 (0.030)
Distance to highways0.032 (0.002)−0.070 (0.004)0.099 (0.005)0.006 (0.005)−0.056 (0.003)0.077 (0.157)
Distance to secondary ways−0.002 (0.001)−0.089 (0.002)0.128 (0.003)0.024 (0.003)−0.051 (0.002)−0.013 (0.126)
Population density0.003 (0.000)0.047 (0.002)0.048 (0.000)0.002 (0.000)0.001 (0.000)−0.107 (0.072)
Urban population density0.010 (0.000)0.000 (0.001)0.009 (0.001)−0.008 (0.001)0.004 (0.000)−0.019 (0.084)
GDP0.041 (0.004)0.030 (0.008)0.131 (0.006)0.019 (0.007)0.005 (0.000)0.003 (0.000)
Constant−1.213−6.672−1.423−3.165−2.412−6.502
Observation numbers165573165573165573165573165573165573
Pseudo R20.1750.4320.2390.1410.1100.315
Table 5. Simulated areas of land use types under different scenarios (Unit: km2).
Table 5. Simulated areas of land use types under different scenarios (Unit: km2).
ScenarioYearCultivated LandForest AreaGrasslandWater AreaBuilt-Up AreaUnused Land
Area % ChangeArea % ChangeArea % ChangeArea % ChangeArea % ChangeArea % Change
Base year2010108,223021,5130807906911022,96401700
Baseline scenario2020106,777−1.3324,38713.35912112.894475−35.2423,0070.1993−45.29
2030107,662−0.5124,39013.376962−13.824557−34.0624,0924.9119715.88
Ecological protection priority2020105,769−2.2625,39718.05907112.284475−35.2423,0550.3993−45.29
203099,023−8.5029,18535.6610,53130.354437−35.7924,6437.3141−75.88
Economic growth202099,298−8.2420,752−3.537189−11.066890−0.3033,71046.7921−87.64
203099,795−7.7820,118−6.486113−24.336897−0.2034,89251.9445−73.53

Share and Cite

MDPI and ACS Style

Hasan, S.S.; Deng, X.; Li, Z.; Chen, D. Projections of Future Land Use in Bangladesh under the Background of Baseline, Ecological Protection and Economic Development. Sustainability 2017, 9, 505. https://doi.org/10.3390/su9040505

AMA Style

Hasan SS, Deng X, Li Z, Chen D. Projections of Future Land Use in Bangladesh under the Background of Baseline, Ecological Protection and Economic Development. Sustainability. 2017; 9(4):505. https://doi.org/10.3390/su9040505

Chicago/Turabian Style

Hasan, Shaikh Shamim, Xiangzheng Deng, Zhihui Li, and Dongdong Chen. 2017. "Projections of Future Land Use in Bangladesh under the Background of Baseline, Ecological Protection and Economic Development" Sustainability 9, no. 4: 505. https://doi.org/10.3390/su9040505

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop