Tools

Intra-annual land use and land cover classification

Our research group has already made good progress in intra-annual LULC classification to determine areas under single cropping (one crop planted in a year, in the Kharif season) or multiple cropping (more than one crop planted in a year, in the Kharif and Rabi seasons), and other more static classes such as forest cover, surface water, and built-up areas. The figure below shows sample output over a few years. These LULC outputs will be used to compute evapotranspiration changes, canopy interception loss, runoff, etc., to construct a time series of groundwater changes.

Building A Vulnerability And Socioeconomic Index

We are examining historic social welfare allocations made across hamlets and villages in the intervention panchayats to identify cases of unfair exclusion and assist the government in formulating vulnerability-proportional welfare allocation to these locations. We sre also obtain accessibility data of social infrastructure such as health facilities and schools from Google or Open Street Maps or data published by the government on the National Data and Analytics Platform (NDAP). Socio-economic factors proxied using a wealth index can be accurately obtained from classifiers trained on satellite-based data. The figure below shows such a sample output from the initial work by our group on poverty mapping. Social welfare expenditures will be obtained from common welfare schemes’ Management Information System (MIS) websites. Detailed health indicators have been estimated through co-variate modelling of coarser-level survey data at the district level, allocated to finer spatial scales.

Landscape-Scale Planning Of Water Structures

Rainwater harvesting site assessment, for example, considers factors such as the soil type, slope, precipitation, soil moisture, land cover and land use. Remote-sensing data products of digital elevation maps and precipitation are now freely available. Machine learning-based land-cover classification algorithms that operate on freely available satellite imagery can provide high-resolution accurate maps, rapidly updatable. Soil moisture can similarly be estimated based on land surface temperature and vegetation indexes. We feel it is, therefore, feasible to conceive of a process whereby such datasets are used to identify candidate rainwater harvesting sites, fairness-based prioritizations are made to ensure the rights of vulnerable communities and participatory methods are then used to obtain community feedback and consensus to demand fund allocation for the identified sites.

Research Paper

Coming soon….

GitHub-Repository

Coming soon….

Community Forest Governance Tools

Geospatial digital applications for forest governance that community members can use to manage their forests better. Forest health is being mapped in high resolution using machine learning on satellite data, which shows vegetation density and open patches resulting from deforestation. Local insights are being added to build rules for the rotational harvesting of forest produce. Hydrological modelling is combined with these maps to recommend locations for new water management structures, contributing to improved forest growth. A geo-tagged incident database is being created to track forest fires, governance violations, corrective measures, and biodiversity, serving as a learning tool for local and other communities and contributing to citizen science databases for climate change.

Research Paper

Coming soon….

GitHub-Repository

Coming soon….

Impact Assessment

The main objective of this research is to quantitatively assess the impact of the water based NRM interventions undertaken by the government scheme Mahatma Gandhi National Rural Employment Guarantee Act (MGNREGA).  

Remote-sensing techniques are being used to monitor pre-post indicators, such as vegetation index, groundwater level, and cropping intensity. This is done to develop a causal analysis framework to understand the change in outcome indicators. 

We envision a tool that will provide an estimate of the impact of NRM assets and a projected impact for prospective NRM sites. Machine Learning (ML) models learned on the past farm ponds can probably be used to predict the site assessment for new prospective farm pond locations, and we will explore this next. 

Research Papers

Coming soon.

GitHub-Repository

Coming soon.

How To’s

Coming soon.

Fairness In Funds Allocation

The objective of this research is to conduct an audit of the fairness in fund allocation in National Rural Employment Guarantee Act (NREGA) scheme at the panchayat level. The focus is on identifying whether there exists any discrimination against Scheduled Castes (SC) and Scheduled Tribes (ST) communities in the provision of employment opportunities under the scheme, despite them being vulnerable and the primary beneficiaries of the scheme. Preliminary experiments showed some instances of discrimination against these communities which motivated us to dive deeper into this.

Additionally, the study seeks to examine whether historically discriminated panchayats that have received lower funding in the past continue to receive less funding over time, or whether changes in political leadership or other factors have resulted in more equitable allocation of funds to these areas. 

Ultimately, the research aims to develop a model to estimate optimal allocation of funds to promote equitable development among the panchayats. The model will take into account various factors such as the socio-economic and demographic characteristics of each panchayat, its developmental needs, and historical funding patterns to ensure that the allocation of funds is fair and just.

Research Papers

Coming soon.

GitHub-Repository

Coming soon.

How To’s

Coming soon.

Landscape Level Planning Of Tree Plantation

This research work focuses on building a participatory tool for assessing the land suitability for tree plantations and further recommend the best suitable tree species for the selected site that caters to the community’s environmental & economical needs. The project uses landscape characteristics like land use land cover, soil type, slope, elevation, proximity to nearest water body and other agro-climatic information of the region to generate the plantation feasibility and species recommendation map. The data for all these characteristics are derived by processing the high resolution open-source satellite imageries using the remote sensing and machine learning techniques.

We are also building a model to assess the plantation’s health over time, by estimating the vegetation density, growth in canopy height and carbon sequestered with time. This would facilitate the farmers, NGOs, government officials and other organizational bodies to analyze the impact of the undertaken NRM tree plantation projects in different places. We envisage to incorporate the learnings from the past plantation projects’ impact in the locality as well as the feedback from the on-ground users constantly into the tool to improvise the tree plantation planning.

Research Papers

Coming soon.

GitHub-Repository

Coming soon.

How To’s

Coming soon.