Assessing land cover land use change (LCLU) in the Central Western Ghats, India is crucial for evaluating ecosystem degradation and environmental processes influenced by climatic and anthropogenic pressures. Despite widespread awareness of the links between biodiversity loss and human wellbeing, the species continue to disappear at an alarming rate. This study investigates the LCLUC dynamics over the past two decades (2002–2023) in Central Western Ghats, spanning an area of 10,419 km2 in Karnataka, India. The Landsat 5 and Landsat 8 satellite imagery served as primary data sources, with Machine Learning Algorithm employed in the Google Earth Engine, a cloud-based platform, to classify images using a Random Forest classifier validated with 17,639 ground training points. The analysis revealed that, dense forest covered largest area (6926.31 ha) followed by open forest (1257.92 ha) and cropland (1195.73 ha). Agroforest and waterbody occupied 249.65 ha and 265.18 ha, respectively, while built-up areas, barren land, and plantation forest constituted 131.38 ha, 83.45 ha and 127.29 ha, respectively. The results indicate a 7.14% reduction in dense forest area, contrasted with a 1.09% increase in agroforestry. The classification achieved an overall accuracy of 94.08% and Kappa index of 0.95. These findings highlight the importance of integrating measures and policies with investment in research and satellite monitoring to curb deforestation in the Central Western Ghats and ensure continuity of Western Ghats in halting degradation and climate change.
Research Article
Geospatial Analysis of Agroforest Landscape in the Central Western Ghats, India, using Google Earth Engine and Machine Learning
https://doi.org/10.21203/rs.3.rs-4836267/v1
This work is licensed under a CC BY 4.0 License
You are reading this latest preprint version
Central Western Ghats
Land Cover Land Use Change
Machine Learning
Agroforestry
Deforestation
Ecosystem Degradation
Climate Change
Land use strategies in agroforest landscapes have profound implications for global endeavors to achieve food security while perpetuating sustainable ecosystem (Foley et al., 2012). Since agricultural revolution, earth’s arable land is being exploited at an accelerating pace in pursuance of nutritional needs of ever-increasing population. The resultant transformation in ecosystem to human subjugated landscapes has been the single largest cause of terrestrial biodiversity loss (Pimm and Raven, 2009). Despite increase in cultivable area and yield, global aspirations to alleviate malnutrition remains elusive (Wu et al., 2014). Consequently, numerous perspectives on optimal landscape management have emerged, focusing on enhancing both agrobiodiversity richness and yield (Phalan et al., 2013).
Agroforestry is a unique land use practice that intentionally integrates woody perennial, herbaceous crops and livestock on a unit piece of land, thereby enhancing productivity and profitability within an agroecosystem (Nair, 1993). Unlike monoculture plantations, agroforestry systems are dynamic, ecologically based natural resource management system that allow farmers to introduce or retain trees within farm and rangeland settings. This integration diversifies and sustains production, providing social, economic and ecosystem benefits (Leakey, 2017). These systems vary in structural complexity, diversity, socio-economic and ecosystem services attributes.
In 2014, India launched the world’s first National Agroforestry Policy, aimed at transforming the lives of rural population by ensuring food security and protecting ecosystem through sustainable means under the theme “Trees for life”. This policy mandates agroforestry promotion on farmlands, expanding tree cover, providing tree insurance, simplifying regulation related to harvesting and marketing of agroforestry products, securing land tenure rights and maintaining a sound base of land records (Chavan et al., 2015). Following this policy, several scientists and organizations began estimating the area under agroforestry. Early estimates noted an agroforestry area of 25.32 Mha (8.2% of total geographical area) (Dhyani, 2014). In contrast, estimates using Bhuvan LISS III satellite imageries by scientists from the Central Agroforestry Research Institute (CAFRI), Jhansi estimated an area of 17.45 Mha of area under agroforestry in India (Rizvi et al., 2014). Surprisingly, most recent estimate by the Forest Survey of India reported area under agroforestry to be 9.50 Mha (2.89%) indicating dramatic decrease in the area and a failure in the practical application of the national agroforestry policy (Anon., 2019).
The discussion on agroforestry today can no longer be confined to land use alone; it must also connect to value chain, consumption and well-being, linking nature to people (Munroe et al., 2019). Agroforestry research and development (R&D) has substantially contributed to the United Nations Millennium Development Goals (MDGs) up to 2014 by enhancing on-farm production and income enhancement for many hungry and impoverished people (Garrity, 2013). In 2015, the United Nations adopted 17 Sustainable Development Goals (SDGs) for transforming world towards sustainability ensuring peace and prosperity for all by 2030 (Goparaju et al., 2020). Agroforestry, in its fifth decade, is driving towards a greater policy synergy connecting technology, landscapes, rights, and markets to restore multifunctionality in the context of the SDGs (Noordwijk et al., 2018). Of 17 SDGs, agroforestry has the significant impact on poverty reduction (SDG 1), hunger alleviation (SDG 2), climate action (SDG 13), biodiversity conservation and sustainable land management (SDG 15). Additionally, it contributes to gender equality (SDG 5), health (SDG 3), access to clean water (SDG 6), sustainable energy solutions (SDG 7) and responsible agricultural production (SDG 12) (Anon., 2020a). To sum up, over 50 per cent of targets, out of 169 fixed in SDGs, are directly related to environment (Singh, 2017).
Karnataka, sixth largest southwestern state in peninsular India, covers roughly 6.3 per cent of total geographical area. Despite its dense population, Karnataka is a mega biodiversity hotspot, featuring a range of ecosystem from wet evergreen to thorn forest and harboring 4700 unique species, including 600 endemics to southern India and 95 species exclusively endemic to Karnataka (Anon., 2020b). The state’s social wellness indicator ranks it relatively high, with a Human Development Index of 0.68, a life expectancy of 70.7 years, a literacy rate of 75.36 per cent and a sex ratio of 973/1000 (Anon., 2020c). The cropping pattern, tree species distribution and food habits are primarily influenced by climate, topography, soil type, crop profitability, irrigation availability and government policies such as subsidies, support price and infrastructure development (Guillerme et al., 2017). Food crops like rice, wheat, maize, sorghum, millets, cereals and pulses dominate Karnataka’s agriculture. In Karnataka state Eastern dry tract, farmers practice growing multipurpose trees on bunds, boundaries and silvi-horti systems, mainly for timber, fuelwood and fodder (Doddabasawa et al., 2020). However, the documentation of agroforestry forms and practices in Central Western Ghats remains an untapped area.
The rapid land use transition in forest and riverine landscapes in Uttara Kannada (Central Western Ghats) witnessed 26 per cent loss of evergreen forest since 1973. This change has been driven by anthropogenic activities such as dam construction, the shift towards monoculture plantation of Teak, Acacia sp. and Eucalyptus sp., agricultural expansion, increased built-up areas, forest-based industries (such as paper mills) and establishment of the Kaiga power plant (Ramchandra et al., 2020). Despite these changes, land use transition in agroforestry systems within the Central Western Ghats or state of Karnataka remains an unexplored area.
Agroforestry, as a stand-alone technology, forms a land-use system, applied after diagnosis and design, participatory research or characterization studies, depending on social, economic and ecosystem constraints (Leakey, 2017). The global knowledge base on agroforestry is rapidly expanding, as evidenced by the quality of scientific publication on various forms and aspects of agroforestry. Although, more than a hundred agroforestry systems have been identified in tropics, the distinction between agroforestry systems and practices often remains unclear and is used interchangeably, referring to forms of land use. Therefore, developing a comprehensive understanding of each land use trajectory is crucial. This includes, identifying the aerial extent of land cover and land use and the components at the farm scale to determine the individual land use changes and the drivers influencing these changes.
Study Area
The Uttara Kannada district, also known as the Central Western Ghats, is located in South West of India between 74º 10’ E and 75º 10 E longitude and 14º 00’ N and 15º 30’ N latitude. The district covers an area of approximately 10,419.10 km2, considered as a unit, in which three distinct bioclimatic zones viz., 140 Km long West Coast, Sahyadri Interior enveloped with dense vegetation and flatten Eastern Plains were identified (Fig. 1).
Land Cover Land Use Change
The study employed the Google Earth Engine (GEE), an open access cloud computing platform, for image collection, supervised classification and accuracy assessment using artificial intelligence and machine learning algorithm. The Landsat data was selected for this study due to extensive temporal coverage, spanning over four decades, and its relatively high spatial resolution of 30 meters, which adequately covers the Western Ghats. The land cover land use trajectories were analysed using Landsat-5 (2001) and Landsat-8 (2023) surface reflectance Tier 1 images collection. The surface reflectance image collections enable cloud masking, using the “pixel qa” band, ensuring high data quality (Table 1). Training areas were identified through field work, and a total of 17,639 ground control points were collected for categorising land cover and land use type using GPS for validation (Fig. 2).
Image data and pre-processing
The image acquisition and pre-processing were conducted for 2002 and 2023 between January to March, corresponding to summer season in the study area, to minimise the weather interference due to low moisture content and cloud cover. A cloud mask was applied to each data set to create a synthetic image with minimal cloud interference. The mask utilized the “pixel qa” band in the surface reflectance collections, effectively removing clouds and shadows and generating cloud free RGB composites. The study area was delineated using a fusion table of district boundaries.
Classification and Accuracy Assessment
The remote sensing analysis involved supervised classification using the Random Forest classifier a decision tree algorithm (Sasikala et al., 2017). Ground truth control points (17,639 points) were used to calibrate high resolution images, generating LCLUC maps for 2001 and 2023 in the Google Earth Pro software. The classification distinguished eight thematic classes such as 1. Agroforestry/Trees outside Forest 2. Cropland 3. Built-up 4. Waterbody 5. Dense forest 6. Open forest 7. Plantation Forest and 8. Barren land (Table 2).
The accuracy assessment was conducted using a confusion matrix to calculate the Kappa index and the overall user and producer’s accuracies (Congalton, 1991).
Where, N represents the total number of classified pixels, A is the sum of diagonal values and B is the sum of the product of row and column totals for each class.
The classified LCLUC images were imported into QGIS in .tiff format to quantify the classes in hectares and to create final land cover land use maps. The LCLUC change graph was developed to assess class variations over the study periods.
Vegetation Change analysis using Normalized Difference Vegetation Index (NDVI)
The analysis of the NDVI values indicated that vegetation reflectance values in 2001 (Fig. 3) were generally lower compared to the observed values of 2023 (Fig. 4). NDVI values ranged from 0.55 to -0.04 in 2023 and from 0.55 to -0.07 in 2001. In 2023, the highest NDVI was recorded for plantation forest (0.55), followed by dense forest (0.53), agroforestry (0.52), open forest (0.41), built-up (0.33), cropland (0.32) and barren land (0.27). In 2001, the NDVI values were highest for plantation forest (0.55) followed by dense forest (0.47), agroforest (0.47), open forest (0.27), cropland (0.25), built-up (0.21) and barren land (0.18). The lowest NDVI values were observed lowest for waterbodies (-0.04 in 2023 and -0.07 in 2001).
Area under land cover land use fraction using Landsat data
In 2001, area under different land use fraction, using Landsat 5 data (30 m resolution) revealed that, dense forest covered the largest area (7,65,496 ha) followed by croplands (1,06,841 ha) and open forest (90,732 ha). The areas under agroforest and waterbody were 14,374 ha and 24,390 ha while built-up, barren land and plantation forest constituted 18020 ha, 11611 ha and 10446 ha, respectively, while built-up areas, barren land and plantation forest occupied 18,020 ha, 11,611 ha and 10,446 ha, respectively (Table 3). The 2001 LCLU imagery depicted croplands mainly on the West Coast and Eastern Plains, while mosaic patches of croplands can be observed in Sahyadri Interior. Built-up areas were less prominent and barren rocky patches were prominently visible along the West Coast. The forest fragmentation on the Southeast was perceived meager, while patches of open forest were predominant along the West Coast. The classification achieved an overall accuracy of 95.02 per cent with a Kappa index of 0.92 (Fig. 5).
In 2023, the area under different land use, using landsat-8 data depicted that, the largest area was still under dense forest (6,91,124 ha), followed by open forest (1,24,273 ha) and cropland (1,20,379 ha). Agroforest and water bodies covered 25,758 ha and 26,593 ha, respectively, while built-up areas, barren land and plantation forests occupied 32,670 ha, 8,187 ha and 12,926 ha, respectively (Table 3). The 2023 LCLU imagery depicted that croplands predominantly expanded on West Coast and Eastern Plains zone while mosaic patches of croplands covered under thick vegetation are visible in Sahyadri Interior. An increased prominence of built-up areas, clustered settlements and fragmentation of dense forests were perceived in Central Western Ghats. The classification achieved an overall accuracy of 94.08 per cent with a Kappa index of 0.95 (Fig. 6).
Changes detection analysis in land cover fractions (2001 to 2023) using Landsat data
Over past two decades, the area under dense forest area declined by 74,372 ha (-7.14%), and barren land by 3,435 ha (-0.33%) in Central Western Ghats. Agroforestry land increased by 11,384 ha (+1.09%), while other land use classes such as open forest (+33541 ha, +3.22%), built-up (+14650 ha, +1.30%) and cropland (+13538 ha, +1.30%) also increased in area. The plantation forest and waterbody areas increased by 2,480 ha (+0.24%) and 2,203 ha (+0.21%), respectively (Fig. 7).
Confusion matrix of land cover land use change (LCLUC) using Landsat data
The tabulated matrix data revealed significant conversion over 20 years (Table 4). Notably, 54,890 ha of dense forest area shifted to open forest, followed by 9,220 ha to cropland, 8,180 ha to agroforest. 4,120 ha to built-up, 2,560 ha to waterbody and 1,270 ha converted to plantation forest. Crop land conversion included 8,100 ha to built-up area, 4,840 ha to agroforest and 1,730 ha converted to water bodies.
Waterbody conversion included 2,240 ha area to cropland and 110 ha area to agroforest. Open forest converted 15,150 ha to cropland, 3,990 ha to dense forest and 1,820 ha area to plantation forest. Barren land conversion included 1,930 ha to cropland and 1,910 ha area converted to built-up areas (Table 4).
Temporal Changes in Land Cover and Land Use in Central Western Ghats
The Central Western Ghats, historically rich in tree diversity covers 6,92,631 ha area of dense forest and 1,25,792 ha of open forest. Cropland occupies 1,19,573 ha, predominantly cultivating paddy, maize, jowar, ragi and sugarcane by farmers in Central Western Ghats. Agroforestry, covering 24,965 ha, chiefly comprised Arecanut based plantation crop combination followed by diverse species homegardens, tropical fruit tree orchards (mainly mango, banana cashew and coconut), commercial timber block plantations (majorly Teak, Mahogany, Red Sanders) etc. Waterbodies covered an area of 26,518 ha, while built-up areas covered 13,138 ha, driven by family fragmentation and population growth. Plantation forest, mainly Tectona grandis and Acacia mangium, covered 12,729 ha which were raised by Karnataka Forest Department, while barren lands, mainly along the West Coast, occupied 8,345 ha land area. These findings align with Koppad and Janagoudar (2014).
Deforestation in Central Western Ghats has reduced dense forest from 73.47% in 2001 to 66.33% in 2023, primarily due to massive deforestation, agriculture encroachment, Arecanut plantation establishment in valleys, clustered settlements and infrastructural development. The conversion of 74,372 ha of dense forest into built-up area, driven by dam construction, road expansion and industrial projects, represents a critical ecological concern in Central Western Ghats (Fig. 7).
The popularity of Arecanut based plantation crop combination led to conversion of 4,840 ha of cropland into agroforest, influenced by attractive market prices. Increased population and consumer price index also expanded built-up areas in Central Western Ghats. The ineffective monitoring resulted in 15,150 ha of open forest converting to cropland. However, afforestation initiatives by Karnataka Forest Department led to the addition of 3,990 ha of dense forest and 1,820 ha of plantation forest in Central Western Ghats (Fig. 8). These observations are consistent with Ramachandra et al. (2016) who reported an evident reduction in evergreen and semi-evergreen forest from 57.31% in 1979 to 32.08% in 2013.
The study utilized the Google Earth Engine platform and Random Forest Classifier to assess land cover land use changes in Central Western Ghats from 2001 to 2023. A significant reduction in dense forest was observed over the past two decades, with ongoing deforestation likely to increase in coming decade. This scenario, analyzed through satellite imageries and considerable ground control GPS points, underscores the government’s inability in fulfilling commitment to halt climate change as pledged at the United Nations Conference on Climate Change (COP). Therefore, there is an urgent need for conservative efforts, investment in research, artificial intelligence, and ground truth data to mitigate deforestation in the Central Western Ghats.
Author Contribution: All authors contributed to conception data collection and designing of the study. Shahbaz Noori contributed by the study with material preparation, data collection, analysis, preparation of first draft of this manuscript. Tabassum H, contributed by assisting in data college of household and ground training points in Central Western Ghats. Shivakumar Inamati contributed by thoroughly guiding, preparation and finalizing of this manuscript. All authors read and approved the final manuscript.
Data Availability: The authors confirm that the data supporting the findings of study are available within the article.
Funding: Directorate of Minorities, Karwar, Government of Karnataka, India.
Conflict of Interest: The authors declare no competing interest.
ACKNOWLEDGEMENT
The authors are grateful to acknowledge the provision of Ph.D. Minority Fellowship provided by Directorate of Minorities, Karwar, Government of Karnataka for providing the financial support to the study.
- Anonymous, 2019, India State of Forest Report. Forest Survey of India (Ministry of Environment, Forests and Climate Change), Dehradun, India, pp. 24-26.
- Anonymous, 2020a, Agroforestry, biodiversity and ecosystem services. Agroforestry Network, Kenya, Africa, pp. 6-7.
- Anonymous, 2020b, Biodiversity status of Karnataka. www.aranya.gov.in/forest.
- Anonymous, 2020c, Social wellness indicators of Karnataka. www.kar.gov.in.
- Chavan, S. B., Keerthika, A., Dhyani, S. K., Handa, A. K., Newaj, R. and Rajarajan, K., 2015, National agroforestry policy in India: a low hanging fruit. Curr. Sci., 108(10): 1826-1834.
- Congalton, R., 1991, A review of assessment the accuracy of classification of remotely sensed data. Remote Sens. Environ., 37: 35-46.
- Dhyani, S. K., 2014, National Agroforestry Policy and the need for area estimation under agroforestry. Curr. Sci., 107: 9-10.
- Doddabasawa., Chittapur, B. M. and Kammar, S. K., 2020, Socio-economic analysis of agroforestry systems in semi-arid tropics. Ind. J. Agrofor., 22(1): 97-102.
- Floreano, I. X. and Moraes, L. A. F., 2021, Land use/land cover analysis (2009-2019) with google earth engine and 2030 prediction using markov-ca in the Rondonia state, Brazil. Environ. Monit. Assess., 2021: 193-239.
- Foley, J. A., DeFries, R., Gregory, P. A., Barford, C., Bonan, G., Stephen, R. C., Chapin, F. S., Coe, M. T., Daily, G. C. and Gibbs, H. K., 2012, Global consequences of land use. Science, 309(5734): 570-574.
- Garrity, D. P., 2013, Agroforestry and the achievement of the millennium development goals. Agrofor. Syst., 61: 5-17.
- Goparaju, L., Ahmad, F., Uddin, M. and Rizvi, J., 2020, Agroforestry: an effective multi-dimensional mechanism for achieving sustainable development goals. Ecol. Quec., 21(2): 1-14.9.
- Guillerme, S., Kumar, B. M., Menon, A., Hinnewinkel, C., Maire, E., Santhoshkumar, A V., 2017, Impacts of public policies and farmers preference on agroforestry practices in Kerala, India. Environ. Manage., 48: 351-364.
- Leakey, R. R. B., 2017, Multifunctional Agriculture - Achieving Sustainable Development in Africa. Academic Press, San Diego, California, USA.
- Munroe, D. K., Batistella, M., Friis, C., Gasparri, N. I., Lambin, E. F., Liu, J., Meyfroidt, P., Moran, E. and Nielson, J. O., 2019, Governing flows in tele-coupled land systems. Curr. Opin. Environ. Sust., 38: 53-59.
- Nair, P. K. R., 1993, An Introduction to Agroforestry. Klgwer Academic Publishers, Dordrecht, Netherlands.
- Noordwijk, M., Duguma, L. A., Dewi, S., Leimona, B., Catacutan, D., Lusiana, B., Oborn, I., Hairiah, K. and Minang, P. A., 2018, SDG synergy between agriculture and forestry in the food, energy, water and income nexus: reinventing agroforestry? Curr. Opin. Environ. Sustain., 34: 33-42.
- Phalan, B., Onial, M., Balmford, A. and Green, R. E., 2013, Reconciling food production and biodiversity conservation: land sharing and land sparing compared. Science, 333(6047): 843-849.
- Pimm, S. L. and Raven, P., 2009, Biodiversity: extinction by numbers. Nature, 403(672): 843-845.
- Ramchandra, T. V., Vinay, S., Bharath, S., Chandran, M. D. S. and Aithal, B. H., 2020, Insights into riverscape dynamics with the hydrological, ecological and social dimensions for water sustenance. Curr. Sci., 118(9): 1379-1393.
- Richards, J. A., 1994, Remote Sensing Digital Image Analysis: An Introduction. Springer-Verlag.
- Rizvi, R. H., Dhyani, S. K., Newaj, R., Karmakar, P. S. and Saxena, A., 2014, Mapping agroforestry area in India through remote sensing and preliminary estimates. Ind. Farm., 63(11): 62-64.
- Sasikala, B. S., Biju, V. G. and Prashanth, C. M., 2017, Kappa and accuracy evaluations of machine learning classifiers. Paper presented in: 2nd IEEE International Conference on recent trends in electronics information and communication technology, 22-24 December, 2017, Bangalore, India.
- Singh, J. S., 2017, Environment: a futuristic view. Curr. Sci., 113(2): 210-217.
- Wu, S. H., Nah, S. L. and Chau, C. F., 2014, Global hunger: a challenge to agriculture, food and nutritional sciences. Crit. Rev. Food Sci. Nutr., 54(2): 151-162.
Table 1. Variables used in the stacked layer and their description
|
Variable |
Description |
Stack layer input |
|
Landsat Bands |
Red, Green, Blue, Near Infrared (NIR) |
Bands for summer season images |
|
Normalised Difference Vegetation Index (NDVI)
|
|
NDVI 2002 and NDVI 2023 summer season images |
|
Slope |
Calculated using GEE slope algorithm(uses the 4 connected neighbours of each pixel to calculate the local gradient) |
Calculated from SRTM data and added as a single band. |
Table 2. LCLUC Classes used and their description
|
Class |
Description |
|
Agroforest / Trees outside Forest (ToF) |
Tree Cover growing outside recorded forest area in rural and urban landscapes |
|
Cropland |
Irrigated and rainfed agricultural landscapes |
|
Builtup |
Buildings, roads, dams and all infrastructural projects |
|
Waterbody |
Reservoirs, Dams, Lakes and Rivers. |
|
Dense Forest |
Recorded forest areas with canopy density of 40% and above |
|
Open Forest |
Recorded forest areas with canopy density between 10 to 40% |
|
Plantation Forest |
Plantations raised by Forest Department |
|
Barren land |
Areas devoid of vegetation (rocky lands) |
Table 3: Change in Land Cover Fraction over past two decades (2002-2023) using Landsat 5 & Landsat 8 in Central Western Ghats
|
Sl. No. |
Land Use |
2002 |
2023 |
Change in Area |
|||
|
Area (ha) |
Per cent |
Area (ha) |
Per cent |
Area (ha) |
Per cent |
||
|
1. |
Agroforestry |
14374 |
1.38 |
25758 |
2.47 |
11384 |
1.09 |
|
2. |
Cropland |
106841 |
10.25 |
120379 |
11.55 |
13538 |
1.30 |
|
3. |
Built-up |
18020 |
1.73 |
32670 |
3.14 |
14650 |
1.41 |
|
4. |
Waterbody |
24390 |
2.34 |
26593 |
2.55 |
2203 |
0.21 |
|
5. |
Dense Forest |
765496 |
73.47 |
691124 |
66.33 |
-74372 |
-7.14 |
|
6. |
Open Forest |
90732 |
8.71 |
124273 |
11.93 |
33541 |
3.22 |
|
7. |
Plantation Forest |
10446 |
1.00 |
12926 |
1.24 |
2480 |
0.24 |
|
8. |
Barren land |
11611 |
1.11 |
8187 |
0.78 |
-3435 |
-0.33 |
|
|
Total Area (ha) |
1041910 |
100 |
1041910 |
100 |
|
|
Table 4: Confusion matrix of land cover land use change using landsat imageries of Central Western Ghats (Area in Sq Km)
|
Ref. class |
Water |
Cropland |
Builtup |
Open Forest |
Agroforest |
Plantation Forest |
Dense Forest |
Barren land |
Total |
|
Water |
219.9 |
22.4 |
0.0 |
0.3 |
1.1 |
0.0 |
0.1 |
0.1 |
243.9 |
|
Cropland |
17.3 |
908.3 |
81.0 |
0.6 |
48.4 |
0.0 |
0.0 |
12.8 |
1068.4 |
|
Builtup |
1.1 |
0.1 |
174.3 |
0.1 |
4.5 |
0.0 |
0.0 |
0.1 |
180.2 |
|
Open Forest |
1.3 |
151.5 |
6.1 |
686.5 |
1.5 |
18.2 |
39.9 |
2.3 |
907.3 |
|
Agroforestry |
0.2 |
9.2 |
4.2 |
0.4 |
119.2 |
1.1 |
7.7 |
1.7 |
143.7 |
|
Plantation Forest |
0.3 |
0.7 |
0.9 |
4.1 |
0.3 |
89.6 |
4.4 |
4.2 |
104.4 |
|
Dense Forest |
25.6 |
92.2 |
41.2 |
548.9 |
81.8 |
12.7 |
6852.4 |
0.1 |
7654.9 |
|
Barren land |
0.2 |
19.3 |
19.1 |
1.8 |
0.9 |
7.6 |
6.7 |
60.5 |
116.1 |
|
Total |
265.9 |
1203.7 |
326.8 |
1242.7 |
257.6 |
129.3 |
6911.3 |
81.8 |
10419.1 |
No competing interests reported.