Expertise
Satellite time series turned into flood, drought and land-cover information that decision-makers can open in a browser.
Suman works with both optical and microwave (SAR) Earth observation data and uses Google Earth Engine daily through its JavaScript and Python APIs. His remote-sensing work has gone into operational products for national and regional agencies: agricultural drought watch systems for Nepal, Bangladesh, Bhutan and Pakistan at ICIMOD, and the Land Cover Change Analytics and fallow-land products for South Asia at RIMES under the World Bank CARE project.
He also does the research side. He co-authored a 2025 Data in Brief paper on time-series flood mapping with Copernicus data in Google Earth Engine for a mountainous region, and a 2024 study of land cover change before and after the Sendai Framework for Disaster Risk Reduction. Most of his projects go past the analysis: the outputs end up in automated pipelines, NetCDF archives served through THREDDS, GeoServer layers and web dashboards that users keep relying on after the project ends.
Maps flood extent through time from Copernicus satellite data in Google Earth Engine, including in mountainous terrain. Co-author of a 2025 Data in Brief paper on this method.
Builds and maintains national agricultural drought watch systems that combine satellite, model and observation data. Systems have been delivered for Nepal, Bangladesh, Bhutan and Pakistan.
Produces land use and land cover change analytics and derived products such as fallow-land maps for regional users in South Asia, with web visualisation built in.
Automated ETL for satellite, model and observation data in NetCDF, HDF, GRIB and GeoTIFF formats using Python, GDAL, CDO and NCO, with QA/QC and PostGIS storage.
Publishes processed archives as standard services through THREDDS Data Server, ncWMS and GeoServer, and builds the Tethys, Vue.js or React dashboards that use them.
Processes UAV surveys into DEMs, point clouds and orthomosaics, as done for a survey in Khotang, Nepal.
Automated Earth Engine workflows that turn satellite time series into regularly updated flood-extent and agricultural-drought layers for your area of interest.
4-8 weeks · fixed-scope project
Land use / land cover maps, change statistics and derived products such as fallow-land layers, packaged so non-specialists can explore them.
4-10 weeks · fixed-scope project
Turn folders of satellite and model outputs into standard NetCDF/THREDDS, WMS and dashboard services that stay up and keep updating.
6-12 weeks, optionally followed by a support retainer · fixed-scope project | retainer
RIMES (Regional Integrated Multi-hazard Early Warning System) · 2024–2025
Regional data products for the World Bank CARE project: a Land Cover Change Analytics product and a South Asia fallow-land product with web visualisation, presented to users at RDAS workshops.
Countries: Nepal, Pakistan
ICIMOD (International Centre for Integrated Mountain Development) · 2019–2022
Tethys-based national agricultural drought watch systems for Nepal, Bangladesh, Bhutan and Pakistan, powered by automated processing of satellite, model and observation data.
Countries: Nepal, Bangladesh, Bhutan, Pakistan
Published research (co-author) · 2025
Peer-reviewed 2025 Data in Brief article (Thapa, Nepali, Shrestha, Sanjel) presenting time-series flood mapping of a mountainous region using Copernicus data in Google Earth Engine.
Where
Loading map…
Tell me what you're working on. A 20-minute call is free and there's no obligation.