Why Banks Need Physical Climate Risk Data: A Playbook for Climate-Smart Lending and ESG Compliance

Agriculture accounts for 6.10% of gross non-performing assets across Indian scheduled commercial banks, the highest of any sector. In H1 FY2025-26 alone, the government wrote off ₹14,122 crore in farm loans due to natural calamities.

For banks with significant agricultural or climate-exposed portfolios, this is not a policy footnote. It is a balance sheet problem, and it is growing every year.

The Problem: Climate Shocks Are Creating a Credit Risk Banks Cannot Ignore

India ranks among the world's most climate-vulnerable nations. The IMF's 2025 Financial Sector Assessment found that in a severe scenario, where climate shocks hit for three consecutive years, agricultural loan default probability could rise by more than 30 percentage points. Expected losses could exceed 25% of total scheduled commercial bank agricultural loans.

This is not a distant projection. Weather-related loan write-offs have grown year on year. As a result, banks face a structural gap. Their credit assessment models were built around borrower financials, collateral, and repayment history. They were not built to account for the flood risk in a borrower's district or the drought probability in a region where a Kisan Credit Card portfolio is concentrated.

This gap has now turned into a regulatory concern. Climate risk in lending has moved from voluntary disclosure territory to a formal compliance requirement.

There are three channels through which physical climate risk reaches bank balance sheets:

• Credit risk: Extreme weather reduces borrower income, driving loan defaults.
• Collateral risk: Floods and droughts can damage or destroy assets pledged against loans.
• Operational risk: Damage to bank branches, infrastructure, and service networks can disrupt operations in climate-affected areas.

What Is Physical Climate Risk Data and Why Do Banks Need It?

Physical climate risk data refers to measurable, location-specific information on climate hazards, including flood frequency, drought probability, heat stress intensity, and cyclone exposure, mapped to specific geographies.

For banks, this data answers a question traditional credit scoring cannot:

How likely is it that this borrower's income or asset base will be affected by weather and climate risks?

The RBI recognised this gap directly. In October 2024, it announced the Reserve Bank Climate Risk Information System (RB-CRIS), a data repository providing standardised hazard, vulnerability, and physical risk datasets to regulated entities. The system launched on 29 May 2025.

This makes physical climate risk data providers, whether public sources such as the India Meteorological Department or private climate-data platforms, increasingly important to how banks conduct climate credit risk assessments.

How Climate Risk Data Applies to Lending Decisions in Practice

Consider a bank managing a Kisan Credit Card portfolio across five drought-prone districts in Vidarbha, Maharashtra.

Using district- and location-level physical risk data, the bank can:

• Identify districts with above-average three-year drought probability.
• Adjust provisioning for climate-exposed loans ahead of the monsoon.
• Require appropriate crop or weather insurance coverage as part of credit renewal in high-risk zones.
• Price risk more accurately instead of applying uniform assumptions across geographically different portfolios.
• Identify concentrations of climate risk across its lending book before losses materialise.

This approach is already emerging within Indian banking. ICICI Bank's FY2025 ESG report describes the use of physical risk assessment for top counterparties in its wholesale lending portfolio, including consideration of the geographic concentration of borrowers in flood- and cyclone-prone regions.

The same logic applies beyond agriculture.

For banks lending to MSMEs in coastal districts, small manufacturers dependent on monsoon-driven supply chains, or businesses operating in heat-stressed urban regions, climate-linked loan default risk is not a theoretical concern. It can directly affect borrower cash flows, asset values, business continuity, and repayment capacity.

What Does Physical Climate Risk Data Enable a Bank to Do?

Action	Without climate data	With climate risk data
Credit appraisal for farm loans	Based primarily on repayment history and financial indicators	Adjusted for location-specific drought, flood, heat, and other hazard exposure
Provisioning for climate-exposed portfolios	Reactive and largely post-disaster	Proactive, informed by hazard probability and portfolio exposure
ESG climate risk scoring of borrowers	Limited or unavailable	Possible using hazard, exposure, and vulnerability data
Climate stress testing	Generic macroeconomic scenarios	Location-specific climate scenarios and hazard pathways
RBI climate disclosure and risk assessment	Primarily qualitative	Supported by quantifiable, location-specific risk data

What Changes When Banks Embed Climate Risk Into Lending?

The RBI's climate-risk framework is pushing banks to move from simply disclosing climate exposure toward identifying, assessing, and managing it.

Banks that integrate climate resilience into their lending workflows can gain three concrete advantages.

First, they can price climate-exposed loans more accurately.

A borrower operating in a high-risk flood zone should not necessarily be treated the same way as an otherwise identical borrower operating in a low-risk area. Location-specific climate information can become an additional risk signal alongside traditional financial and behavioural indicators.

Second, banks can reduce surprise losses.

If a portfolio is heavily concentrated in districts facing recurring drought or flood exposure, identifying that concentration before an extreme event gives the bank an opportunity to adjust provisioning, portfolio limits, insurance requirements, or credit strategy.

Third, banks can strengthen their ESG and climate-risk reporting.

Instead of relying primarily on qualitative statements, lenders can use physical hazard, exposure, and vulnerability data to quantify climate risks across borrowers and portfolios.

Where Parametric Insurance Fits In

Parametric insurance adds another layer of resilience.

A farmer borrower covered by parametric weather insurance can receive a predetermined payout when an objectively measured trigger—such as rainfall, temperature, or another defined weather variable—is breached.

This can protect liquidity after a climate shock without requiring a traditional loss-assessment process.

For a lender, that matters because borrower resilience and repayment capacity are closely connected.

Consider two otherwise similar agricultural borrowers:

• Borrower A has no climate-risk protection.
• Borrower B has parametric weather coverage linked to a defined climate trigger.

If a severe drought occurs and the trigger is breached, Borrower B receives a payout that can help stabilise household or farm cash flows.

This does not eliminate credit risk. However, it can potentially reduce the severity and duration of climate-driven financial stress.

For financial institutions looking to reduce agricultural NPA risk, parametric insurance can therefore be considered not only as an insurance product, but also as a portfolio resilience mechanism.

Limitations for Banks Using Climate Risk Data

Integrating physical climate risk data into lending is not simple. Banks face several practical constraints.

Data Quality and Availability Remain a Central Challenge

RB-CRIS is designed to improve access to standardised climate-risk information, but banks still need to integrate data from multiple sources, geographic scales, and formats.

ICICI Bank's FY2025 ESG report acknowledges that available climate data can be fragmented, available in varied formats, and inconsistent in frequency and units.

A bank therefore needs more than a climate dataset. It needs a usable climate-risk intelligence layer that can connect hazard information with its actual lending exposure.

Historical Data Alone Is Not Enough

Historical weather observations are essential for understanding long-term patterns, but they do not necessarily represent future climate conditions.

As climate patterns shift, banks need to combine:

• Historical observations
• Recent weather data
• Climate projections
• Forecasting
• Hazard probability
• Location-level exposure
• Borrower and portfolio information

This combination creates a more useful climate credit risk assessment than relying on historical averages alone.

Loan-Level Geolocation Is Critical

A bank cannot accurately assess physical climate risk if it does not know where its borrowers, collateral, branches, or financed assets are located.

The IMF's 2025 FSAP highlights the importance of loan-level geolocation data in understanding climate-related tail risks.

For example, knowing that a bank has ₹500 crore of agricultural exposure in a district is useful. Knowing that ₹120 crore of that exposure is concentrated within specific flood- or drought-prone areas is substantially more actionable.

Basis Risk Still Applies

Even sophisticated climate models cannot perfectly capture every hyper-local event.

A flash flood may affect one block while leaving a neighbouring block largely unaffected. A district-level drought indicator may also conceal differences in local rainfall, soil conditions, irrigation access, or crop exposure.

This creates a form of geographic aggregation risk.

The more granular the data, the more useful it becomes for credit decisions—but also the more important data quality and validation become.

Integrating ESG Climate Risk Scoring Into Credit Decisions

To properly integrate ESG climate risk scoring with climate-linked loan default risk, banks need the right internal capabilities.

This includes:

• Climate-risk expertise
• Geospatial analytics
• Data engineering
• Portfolio-level risk modelling
• Scenario analysis
• Risk governance
• Staff training

For smaller banks and regional rural lenders, building all of these capabilities internally can be difficult.

This is where specialised physical climate risk data providers and climate-risk technology platforms can play an important role by converting complex climate datasets into decision-ready risk indicators.

A Practical Playbook for Climate-Smart Lending

For banks beginning their climate-risk journey, the process does not need to start with a complete overhaul of the credit system.

A practical approach can be built in five stages:

1. Map the portfolio

Identify where borrowers, collateral, branches, and financed assets are located.

2. Map physical hazards

Overlay portfolio locations with relevant hazards such as flood, drought, extreme heat, cyclone, and rainfall variability.

3. Quantify exposure

Determine which sectors, geographies, borrower segments, and loan books are most vulnerable.

4. Integrate risk into decisions

Use climate indicators in credit appraisal, portfolio monitoring, provisioning, stress testing, insurance requirements, and risk-based pricing where appropriate.

5. Monitor and update

Climate risk is dynamic. Banks should periodically refresh hazard data, forecasts, borrower exposure, and portfolio concentrations rather than treating climate assessments as a one-time exercise.

This creates a progression from climate data → climate intelligence → credit decision → portfolio resilience.

Wrapping Up: The Regulatory Direction Behind Climate Credit Risk Assessment

Climate risk in lending has moved from a voluntary consideration toward a core financial-risk issue for Indian banks.

The RBI's climate-risk initiatives, combined with the launch of RB-CRIS, signal a broader direction: physical climate risk data will increasingly become part of how financial institutions understand credit, portfolio concentration, stress scenarios, and resilience.

Agriculture remains particularly exposed, with the IMF noting that agriculture accounts for approximately 13% of scheduled commercial bank loans and more than 66% of regional rural bank loans.

But the opportunity extends well beyond agricultural lending.

MSMEs, infrastructure, housing, supply-chain finance, and other climate-sensitive portfolios can all be affected by physical hazards.

The banks that begin integrating location-specific climate intelligence into their risk architecture today will be better positioned to identify vulnerable portfolios, design resilience mechanisms, and make more informed lending decisions as climate risk becomes increasingly material to financial performance.

Ready to Learn More About Climate Risk Data and Its Impact on Loans?

Parametric insurance is quietly becoming part of how forward-looking lenders manage climate-exposed portfolios.

Explore how climate intelligence, physical risk assessment, and parametric insurance can work together as practical tools for building more resilient lending portfolios across agriculture and other climate-exposed sectors.