Over the past few years, the weather across India’s farmlands has shifted from unpredictable to unsettlingly erratic. One season brings a relentless heatwave that scorches crops before they can grow. The next brings erratic rains that drown freshly sown seeds altogether. In 2022, an intense heatwave across Punjab and Haryana from March to May reduced wheat yields by 10% to 15%. With more than 357 million tons of annual foodgrain production that support nearly 810 million National Food Security Act (NFSA) beneficiaries, climate shocks are more than agricultural events. These pose systemic risks to India’s procurement system, buffer stocks, and food price stability.

Farming has always been risky. Yet for many, it has now become an even higher-stakes gamble. An article published in Agricultural Reviews by the Indian Council of Agricultural Research (ICAR) warns that without adaptation, national productivity could fall by 40% by the 2080s. While better seed varieties and irrigation systems can help, they cannot compensate for the absence of timely, precise information reaching the farmer who needs to act on it.

Amid intense climate volatility, India’s digital initiatives and innovations during the past decades have emerged as a beacon of hope. Initiatives, such as fertilizer delivery through the Direct Benefit Transfer (DBT) mechanism and the creation of foundational registries under AgriStack have begun to transform Indian agriculture. The nationwide soil health mapping through the Soil and Land Use Survey of India (SLUSI) and the national Soil Health Card (SHC) program have strengthened the foundation of digital agriculture.

Together, these efforts address long-standing challenges in agriculture and help establish a strong digital backbone. This foundation also creates new opportunities for innovation, such as Krishi DSS, a geospatial decision support system, and Bharat VISTAAR. The latter is an AI-powered digital public infrastructure launched recently by the Government of India. It integrates agricultural data, advisories, market information, and government programs to help farmers make better farming decisions.

The climate gap in India’s digital agriculture foundations

Each of these digital interventions was designed to address a specific challenge and largely functioned in isolation. These verify identity, deliver inputs, and manage subsidies. However, none were built with climate considerations in mind. As a result, the data remain siloed, while the outcomes generated by these systems lack a climate lens. A farmer may know the soil’s nutrient profile, receive a timely fertilizer subsidy, and be registered in a crop database. Yet the absence of integrated early warning systems prevents the farmer from being able to act when a heatwave or unseasonal rain threatens the harvest.

Figure 1: India’s agri-digital systems lack real-time climate integration

The above agri-digital systems have laid the rails. The next step is to create and integrate a climate resilience layer that works in tandem with existing systems and digital public infrastructure. This integration would transform agricultural advisory into climate-smart advisories that can be disseminated to farmers through multiple channels. Without a dedicated “climate layer” embedded within India’s digital agriculture ecosystem, the risk of farmers being exposed to climate hazards increases significantly.

Therefore, the climate-resilient agricultural system (CRAS) illustrates how a network of agricultural data, when combined with a climate intelligence layer, can enable the delivery of climate-specific advisories. Embedding climate analytics into farm-level decision-making enables the generation of hyper-local, data-driven advisories that can be accessed through any farmer-facing platform.

The CRAS can use Bharat VISTAAR and other such open networks and platforms to expand its reach. This will enable interoperable dissemination of these climate analytics and advisories across multiple digital platforms. It will ensure that actionable insights reach farmers through any farmer-facing platform. At its core, the system consist of the following capabilities:

1. Using existing data sources: The CRAS will draw upon agriculture and climate-related datasets available across existing government and partner systems, fertilizer use, and subsidy records under Direct Benefit Transfer (DBT). Weather forecasts from the India Meteorological Department (IMD) will also serve as key inputs. Satellite imagery will help monitor crop stress, vegetation health, and heat exposure. The CRAS will not centrally aggregate all datasets. Instead, it will use relevant data from these pipelines and interoperable registries, such as those under AgriStack, to generate climate-aware insights that can support agricultural planning and localized advisory services.
2. Climate analytics: The CRAS will integrate existing climate advisory models with climate and agricultural datasets to generate localized climate intelligence. These advisory models include the ICAR’s InfoCrop Model and the . An additional artificial intelligence (AI) layer in the farmer-facing applications will help identify patterns across weather, soil, and crop data. This will enable personalized use cases for farmers, such as early warning systems, hyper-local advisories, and climate-resilient recommendations on risks, which include droughts, pest outbreaks, and extreme weather events.
3. Bharat VISTAAR layer: The CRAS climate analytics layer will generate insights that can be shared across agriculture service providers through Bharat VISTAAR. Multiple applications and service providers within the digital agriculture ecosystem will access these climate-informed insights directly.
4. Farmer-facing dissemination: Farmer-facing platforms, such as Bihar Krishi, MahaVISTAAR, and the TNAU AgriTech Portal can translate these insights into personalized, localized advisories and services. Farmers may receive hyper-local weather alerts, crop planning recommendations, pest and disease warnings, and supply chain notifications. They may also receive financial triggers, such as subsidy eligibility and credit support, through SMS, WhatsApp, mobile applications, or extension networks.

The CRAS will not reinvent the wheel but connect the spokes. The digital systems are already in place. The challenge and opportunity are to make them speak to each other in a climate-aware language.

Figure 2: Data architecture of CRAS

Recent efforts by the government to connect platforms, such as Bharat VISTAAR with AgriStack, indicate a move toward greater interoperability in digital agriculture. However, we must distinguish between roles. The CRAS functions as a climate intelligence and advisory layer. It will integrate weather data, climate models, and agricultural system data to generate climate-informed advisories that strengthen farm-level decision-making. It can also be imagined as a node for other open networks.

As more platforms and datasets connect to this network, Bharat VISTAAR can create a compounding network where improved data exchange and wider participation continuously enhance the precision, reach, and usefulness of climate-informed agricultural advisories.

Use cases: CRAS in action

Once operational, the CRAS could reshape decision-making for farmers, researchers, and financial institutions alike. A successful CRAS would enable the following use cases:

1. Climate-smart crop planning: Once enabled, the CRAS would integrate farm-level data with climate intelligence to generate localized, climate-resilient crop recommendations by combining soil profiles, crop histories, and climate forecasts. It would account for soil moisture, rainfall variability, and monsoon onset. This would guide farmers away from high-risk, water-intensive crops and deliver real-time advisories on irrigation, fertigation, and pest management. Such steps would help reduce yield losses, stabilize incomes, and align farm decisions with evolving climate and market signals.
2. Climate-informed financial decision-making: Once implemented, CRAS would enable financial institutions to make more informed decisions by using its unified data and climate intelligence layers. This approach would support more targeted and responsive financial investments in the context of climate shocks. Access to farm-level data, such as crop type, soil quality, and land ownership, combined with insights from the climate intelligence layer, could enable institutions to develop more accurate risk profiles of farmers and significantly reduce uncertainties
3. Localized climate and pest early warning system (EWS): The CRAS architecture could aggregate data from multiple sources to enable a localized EWS that issues hyper-local alerts for extreme events and pest outbreaks. Paired with short-term, actionable advisories, these alerts would allow farmers to take preventive action and reduce losses before impacts occur. Simultaneously, the CRAS would anticipate changes in demand, availability, and product types required across regions to enable companies to optimize their supply chains. Furthermore, farmer reports collected before and after events would strengthen CRAS as a resilient and trusted system.

Learning models from around the globe

While the CRAS will take time to become operational, several promising efforts already exist. In Telangana, the Data in Climate Resilient Agriculture (DiCRA) platform, developed in collaboration with the UNDP, uses satellite data and open-source AI to provide location-specific climate risk advisories. Its modular, plug-and-play design has allowed it to expand to other Indian states and into pilot programs in Latin America.

Beyond India, the Agricultural Climate Resilience Enhancement Initiative (ACREI) is another initiative led by the World Meteorological Organization in collaboration with the Food and Agriculture Organization (FAO). It has been rolled out in Ethiopia, Uganda, and Kenya, and reaches approximately 1,800 farming households. It combines integrated early-warning systems, real-time climate data, and on-ground farmer training.

Conclusion

Over time, the CRAS could expand beyond crop agriculture. Linking it with the National Digital Livestock Mission (NDLM) and forest and vegetation datasets could enable a more comprehensive view of emissions and carbon sinks across the agriculture, forestry, and other land use (AFOLU) sector. Together, this integration would reduce information asymmetries, strengthen climate risk assessment, and help unlock greater public and private investment in climate-smart agriculture.

A looming crisis stares down at India. As the world’s most populous nation and one of the fastest-growing economies, the country faces intensifying climate risks that directly threaten lives, livelihoods, and development gains. Over the past five years, extreme climate events have affected more than 75% of India’s districts and damaged more than 36 million hectares of crops. These numbers underscore the sheer scale, frequency, and systemic nature of climate shocks.

Despite this growing exposure, several constraints hold back India’s ability to plan for, finance, and implement climate adaptation and resilience. Climate systems and decision-support mechanisms are fragmented, siloed, and slow, with limited integration across sectors, such as agriculture, water, urban development, health, and finance. Even where sector-specific systems are available, they often lack a coherent climate risk and resilience lens. The result is reactive responses rather than anticipatory, risk-informed action. Such structural weaknesses can directly hurt effective climate finance.

In the section below, we discuss the challenges with a data lens.

1. Fragmented and outdated climate data ecosystem: Climate data in India is scattered across sectors and stored in inconsistent formats. The Ministry of Agriculturecollects land-use data through the Annual Crop Production Survey, while the Forest Survey of India conducts biennial assessments of forest cover. The Central Pollution Control Board monitors industrial emissions. Each of them uses different methodologies, timelines, and standards. This makes cross-sectoral analysis nearly impossible and creates blind spots.
2. Delayed emission mapping and reporting: Current greenhouse gas (GHG) inventories and biennial update reports (BURs) rely on episodic submissions rather than continuous monitoring. For instance, India’s third biennial update report(BUR-3), submitted to the United Nations Framework Convention on Climate Change (UNFCCC) in 2021, reported emissions data from 2016, a gap of five years.
3. Siloed vulnerability and climate action tracking: Climate risks across sectors are currently assessed in isolation, with limited integration between ministries and programs.. For example, the Pradhan Mantri Fasal Bima Yojanaor crop insurance scheme could benefit from stronger real-time integration and data interoperability with the India Meteorological Department’s early warning systems and the National Disaster Response Force’s preparedness programs.
4. Weak integration of climate data in budgeting and planning: At the grassroots level, recent efforts, such as the Climate Resilience Planning initiativeby the Ministry of Panchayati Raj and the UNDP, seek to integrate climate risk data into Gram Panchayat Development Plans (GPDPs). However, many panchayats still lack the capacity or tools to fully operationalize this data.

A unified digital public infrastructure (DPI) for climate can provide a transformative pathway to build resilience, enhance credibility, and position India as a global leader in climate action. Once operational, the DPI for climate could have an impact comparable to that of the India stack, which comprises Aadhaar-based eKYC, e-sign, and UPI in the financial sector. It could also have a similar impact to what the under-development AgriStack is expected to achieve for agriculture through the creation of reliable farmer- and farm-level data.

Introducing the climate stack: A unified climate intelligence system

The climate stack represents a paradigm shift from fragmented datasets to a unified, dynamically linked infrastructure that serves as a single source of truth on climate intelligence. Rather than create another database, the climate stack brings together a collection of core interoperable registries that address the following fundamental questions:

1. Emission registry: Which areas emit greenhouse gases, and how much?
   This registry tracks greenhouse gas emissions across space and time to provide granular visibility into emission sources and trends.
2. Green assets and sequestration registry: Where are the carbon sinks, and what solutions are available?
   This registry maps India’s natural capital, such as forests, wetlands, grasslands, and agricultural soils, which sequester carbon and provide ecosystem services.
3. Climate vulnerability registry: Which areas and communities are most at risk?
   This registry assesses exposure, sensitivity, and adaptive capacity across India’s diverse landscapes and populations. It maps adaptation and mitigation actions, such as irrigation expansion, resilient seeds, watershed projects, afforestation, and renewable energy uptake, and also tracks their impact on reducing vulnerability over time.

The emissions profile spans four major sectors – energy, agriculture, waste, and industrial processes and product use (IPPU), each with distinct characteristics and data needs. In 2020, India emitted 2,958,589 Gg of CO₂-equivalent GHGs from the following sectors; thus, the climate stack should serve these four critical sectors:

Share of sectors in emissions as per BUR-4 MoEFCC, 2011
India BUR-4.pdf

Why should agriculture be the starting point?

1. Agriculture is a major source of emissions, a sector of high vulnerability, and a domain with a high potential for mitigation.
   • A major source of emissions:Agriculture contributes 13.72% of India’s total GHG emissions (BUR-4), primarily from enteric fermentation, rice cultivation, and fertilizer use. Globally, the agriculture, forestry, and land-use (AFOLU) sector accounts for 22% of emissions (IPCC AR6).
   • Highly vulnerable to climate change:More than 59% of India’s workforce depends on agriculture. Between 2016 and 2021, climate extremes damaged 36 million hectares of crops, which caused losses worth USD 3.75 billion. Yield losses could rise 10–25% by 2050 without adaptation (ORF).
   • High mitigation and adaptation potential:Climate-smart practices can deliver 30%+ of India’s total mitigation potential by 2030. Soil carbon enhancement and restoration of degraded farmlands offer some of India’s lowest-cost mitigation opportunities.
2. Agriculture already has data systems that map directly onto the three climate stack registries. These operational data systems can be integrated immediately.
   • Emissions registry: The major emission categories of agriculture—fertilizer use (N₂O), rice cultivation (CH₄), enteric fermentation and manure management (CH₄/N₂O), and crop residue burning—all have existing digital data sources. These include AgriStack, the Soil Health Card, DBT fertiliser systems, and the livestock stack. Together, these categories represent more than 80% of agricultural emissions. This makes agriculture the most natural starting point for the proposed emissions registry.
   • Sequestration registry: Data for carbon sinks is more complex to source, though some are available readily in existing systems. These include the Forest Survey, which captures forest biomass, canopy density, and carbon stock, and the Soil Health Card, which tracks soil organic carbon for croplands. The AgriStack further provides data on cropland area, land use, and residue cycles. Since croplands and forests form the bulk of India’s terrestrial carbon sinks, governments can digitize and geotag datasets to build the sequestration registry.
   • Vulnerability and climate action registry: Agriculture produces the richest hyperlocal climate-risk signals. These signals cover exposure measured through maximum and minimum temperatures, and sensitivity measured through annual rainfall and disaster proneness. Adaptive capacity is captured through irrigation access, extension reach, and credit penetration. These align directly with the exposure–sensitivity–adaptive capacity framework used by the ICAR.

What should the climate stack’s data architecture look like?

The climate stack is conceived as a federated digital public infrastructure that unifies India’s climate-relevant datasets without centralizing them. The architecture described below represents the climate stack’s overall design logic and is intended to remain consistent across sectors. However, as agriculture, forestry, and land use form the natural starting point, the illustration below presents the same architecture through an AFOLU lens. This version does not alter the core design. It simply maps the data ingestion layer to AFOLU-relevant systems to show how existing datasets flow into the federated registries. These systems include agriculture databases, remote sensing platforms, climate models, and disaster management systems.

Data architecture for climate stack (for AFOLU sector)

The climate stack will not replace existing systems. It will link them through shared standards and interoperable APIs. A data aggregation and standardization layer harmonizes diverse data sets using common taxonomies, identifiers, metadata standards, and quality protocols. This enables integrated analysis across satellite data, sectoral records, weather feeds, and disaster information. These harmonized inputs feed into three federated core registries, namely, emissions, sequestration, and vulnerability. Each of these answers a fundamental climate question and leaves raw data with respective custodians.

A dedicated governance layer ensures the stack is trusted, secure, and institutionally sound. It draws on existing legal and policy frameworks for data protection, privacy, consent-based sharing, interoperability, and stewardship. Clear data-sharing agreements and accountability mechanisms provide confidence across ministries, states, and stakeholders.

The stack unlocks value through clearly defined use cases. It enables high-integrity emissions and sequestration reporting, and real-time and hyper-local climate advisories. It also integrates climate intelligence into local and district planning and ensures the development of climate-resilient financial and insurance products. It supports mitigation and adaptation objectives as it structures climate intelligence into decision-ready formats.

This data architecture serves a diverse set of users. Governments can gain reliable intelligence for reporting and policy, while farmers receive actionable advisories. Local bodies can embed climate risk into planning, and private sector and financial institutions can build resilience-linked products. This is how the climate stack will shift India from fragmented data systems to coordinated, data-driven climate action.

The climate stack offers India an integrated, future-ready approach to climate action, with agriculture as the natural entry point. Its design is intentionally modular, and as India advances digitalization across sectors, the stack can expand to cover energy, waste, and IPPU. Energy is already the next frontier, with early groundwork underway for an energy stack.