ICAR – Indian Agricultural Statistics Research Institute

Reading the numbers nature leaves behind.

We develop statistical and computational methods for ecological and environmental data — from species distributions and biodiversity indices to climate variability and its bearing on Indian agriculture.

Explore our research Meet the team

About the lab

Statistics for landscapes, species and climate.

The Statistical Ecology and Environmental Statistics group at ICAR-IASRI works at the interface of ecology, environmental science and modern statistical methodology. Our work supports researchers, conservation planners and policymakers who need rigorous, data-driven answers to questions about species, habitats and a changing climate.

We build and apply methods for spatial and spatio-temporal modelling, biodiversity assessment, population dynamics and time series forecasting — tailored to the messy, often sparse data that ecological and environmental studies produce.

The group's scope extends into statistical and quantitative genetics as well: genomic selection, genome-wide association studies (GWAS) and genotype × environment interaction, alongside machine learning approaches for high-dimensional ecological and genomic data.

Lab members

Research areas

ICAR-IASRI

Host institute, New Delhi

Research areas

Where we focus our work.

A cross-section of the methodological and applied problems the lab works on.

Species distribution & habitat modelling

Statistical models linking occurrence and abundance records to environmental covariates, used to map habitat suitability and predict range shifts.

Biodiversity & ecological indices

Methods for estimating species richness, diversity and community structure from field survey data, including approaches for imperfect detection.

Spatial & spatio-temporal statistics

Geostatistical and areal models for environmental surfaces that change across both space and time, drawing on remote sensing and GIS data.

Climate variability & agricultural risk

Time series and extreme-value methods for characterising climate variability and its downstream effects on crops, pests and yield stability.

Population dynamics

Demographic and stochastic process models for tracking how populations of plants, animals and pests grow, decline and interact over time.

Wildlife & forestry statistics

Sampling design and estimation methods for forest inventories, camera-trap studies and wildlife population surveys.

Time series analysis & forecasting

Classical and modern forecasting methods for ecological, climatic and agricultural time series, including nonlinear and non-stationary processes.

Machine learning for ecological & genomic data

Predictive and pattern-recognition methods — from ensemble learning to deep learning — applied to high-dimensional ecological and genomic datasets.

Statistical genetics

Quantitative and population genetics methods for dissecting the genetic architecture of traits, including heritability and variance component estimation.

Genomic selection

Whole-genome prediction models that use marker data to estimate breeding values and accelerate genetic gain in crop and livestock improvement.

Genome-wide association studies (GWAS)

Statistical methods for linking genetic markers to phenotypic traits across the genome, including mixed models that account for population structure.

Genotype × environment interaction

Models for understanding how genotypes perform differently across environments, supporting stability analysis and site-specific variety recommendation.

Team