About LEAF DSS
A GIS-enabled Decision Support System for evidence-based planning of agroecology initiatives across Assam, India.
LEAF (Landscape Evaluation and Assessment Framework) is a GIS-enabled Decision Support System that supports evidence-based planning for agroecology (AE) initiatives including livestock, organic, natural farming and integrated farming systems. LEAF collates a multi-dimensional dataset at state, district, block, and cluster (group of villages) level, drawing on over 120 indicators organised into seven data categories. Together, these categories allow LEAF to represent not only biophysical suitability, but also the enabling conditions required for successful adoption and scaling of AE pathways.
To ensure replicability and scalability, LEAF relies on freely available and open data sources wherever possible, so that any state or implementing agency can adopt the tool without dependence on proprietary datasets.
LEAF enables users to identify suitable areas and landscapes — including districts, blocks, or village clusters — by generating feasibility and suitability assessments for agroecology-aligned interventions, including Organic, Natural, and livestock-based approaches.
It is designed to support government schemes and programmes such as PKVY (Paramparagat Krishi Vikas Yojana) and DAY–NRLM (Deendayal Antyodaya Yojana – National Rural Livelihoods Mission), the National Mission on Natural Farming and Assam’s Mukhya Mantri Mahila Udyamita Abhiyan (MMUA) and UDAYINI, which envisage cluster-based strategies to promote livestock, organic and natural farming in a structured, scalable, and market-linked manner.
At the core of LEAF’s decision-making layer is a curated set of 120+ variables drawn from open, government-published datasets — Mission Antyodaya, Livestock Census, IMD, MMUA scheme, SRLM programme data, remote sensing, and soil health sources. These variables are organised into seven thematic dimensions and are pre-assigned as critical variables with recommended ranges based on literature review and consultations for each supported intervention type.
However, the system is configurable: planners can adjust which variables are included, set preferred ranges (e.g. minimum cattle density, maximum groundwater exploitation), and modify weights to reflect local programme priorities or evolving policy guidance. For any selected area — a district, a block, or a cluster of villages — LEAF compares actual values against these configured thresholds and generates a composite suitability profile. Each variable contributes a scored signal: whether it is in the preferred range (strength), close to it (marginal), or significantly below it (gap requiring attention). This produces a transparent, indicator-by-indicator picture of where an area stands — not a black-box score, but a structured diagnostic that practitioners can read, interrogate, and act on.
A summary dashboard view gives planners a 5–6 line snapshot with all of the data of the area’s key strengths, critical gaps, and headline recommendations — suitable for presentations, review meetings, and quick prioritisation. A detailed report supported by an AI recommendation engine, working with this data and those background references, provides full indicator values, dimension-wise scores, gap analysis, and specific recommendations — grounded in both the quantitative data and the policy background documents that define what “suitable” means for that intervention type.
LEAF operates across four spatial scales, allowing planners to move seamlessly from state-wide prioritisation down to individual village cluster formation. Each level of the platform presents the same underlying data at a different resolution, with outputs calibrated to the decisions made at each scale.
Assam’s diverse agro-climatic zones — from the Brahmaputra floodplains to the Barak valley and hilly districts — require highly localised planning. The Assam State Rural Livelihood Mission (ASRLM) works with over 27 lakh SHG households across 34 districts. LEAF enables ASRLM teams to spatially identify where Organic Farming, Natural Farming, or Livestock clusters are most suitable and back those decisions with data-driven evidence for approvals, convergence planning, and reporting.
LEAF is designed to support ASRLM teams in making clear, evidence-based, and scalable planning decisions for agroecological initiatives. It can be used across the planning cycle — from identifying priority geographies to generating communication-ready outputs for approvals and stakeholder coordination.
