AVHRR High-Latitude Precipitation Retrieval
Development and diagnostic evaluation of AVHRR-based precipitation retrievals for high-latitude regions where conventional satellite precipitation estimates remain uncertain.
Remote sensing and precipitation retrieval
Kwabena Kingsley Kumah
Postdoctoral Research Associate
Department of Hydrology & Atmospheric Sciences, University of Arizona
Remote sensing scientist developing and evaluating satellite-based precipitation and cryosphere records.
About
Bio
Bio
Dr. Kwabena Kingsley Kumah is a remote sensing and precipitation retrieval scientist specializing in satellite algorithm development, precipitation-product evaluation, geospatial machine learning, uncertainty analysis, and climate-data records.
He is currently a Postdoctoral Research Associate in the Department of Hydrology & Atmospheric Sciences at the University of Arizona, where he contributes to NASA-supported precipitation research relevant to GPCP, IMERG, AVHRR-based high-latitude precipitation retrieval, ocean precipitation evaluation, polar precipitation benchmarking, and satellite/reanalysis product validation.
His current work includes AVHRR infrared retrievals, diagnostic evaluation of satellite precipitation products, uncertainty-aware validation, and applications in ocean and polar precipitation. This research addresses environments where precipitation is difficult to observe directly and where satellite records must be carefully interpreted.
Dr. Kumah completed doctoral work at the University of Twente / ITC, where he developed high-spatiotemporal rainfall estimation methods using commercial microwave link attenuation and MSG satellite observations. That earlier work focused on improving rainfall information in data-sparse regions in Africa, where gauges and weather radar networks are often limited.
Across these efforts, his broader motivation is to improve precipitation information where observations are sparse, uncertain, or difficult to obtain.
Education
Ph.D., GIS & Earth Observation, University of Twente / ITC, 2022
Dissertation: High-Spatiotemporal Resolution Rainfall Estimation from Satellite and Commercial Microwave Link Data
M.Sc., GIS & Earth Observation, Water Resources & Environmental Management, University of Twente / ITC, 2016
B.Sc., Environmental Science, University of Cape Coast, Ghana, 2012
Research topics
Projects
Featured Projects
Selected work across satellite retrieval, validation, and data-sparse rainfall monitoring.
AVHRR Limb-Darkening Correction
A physically informed correction framework for reducing scan-angle-related brightness-temperature bias in AVHRR infrared observations used for precipitation retrieval.
Integrated Ocean Precipitation Evaluation
A multi-reference validation framework comparing GPCP, IMERG, ERA5, and MERRA-2 with PAL, moored buoys, atolls, and OceanRAIN across daily to climatological scales.
Antarctic Snowfall Mass-Budget Benchmarking
A physically constrained assessment of Antarctic precipitation using GRACE-based storage change, ice discharge, basal melt, and sublimation inputs across IMBIE drainage basins.
GMASI Snow and Ice Cover Extension
A machine-learning extension of the Global Merged Analysis of Snow and Ice record back to 1980-1987 using ERA5-derived surface variables.
CML-Satellite Rainfall Retrieval for Africa
Rainfall detection and mapping methods combining commercial microwave link attenuation with Meteosat cloud-top observations for data-sparse regions.
Research
Research Themes
1
Improving satellite precipitation retrievals
Dr. Kumah works on AVHRR-based high-latitude precipitation retrievals, limb-darkening correction, machine learning, and diagnostic evaluation relevant to long-term precipitation records such as GPCP and IMERG. This work focuses on improving the physical consistency and interpretability of satellite precipitation information in challenging environments.
2
Evaluating precipitation products across scales
His validation work compares products including GPCP, IMERG, ERA5, and MERRA-2 with independent references such as PAL, moored buoys, atolls, OceanRAIN, and Antarctic mass-budget constraints. The emphasis is on uncertainty-aware validation across daily, seasonal, and climatological scales.
3
Expanding rainfall information in data-sparse regions
Earlier research combined commercial microwave link attenuation with MSG SEVIRI cloud-top observations to support rainfall detection and mapping in Sub-Saharan Africa. This work reflects the broader challenge of precipitation monitoring where gauge and radar networks are limited, including rainfall-monitoring relevance for Ghana and other African regions.
Publications
Selected publications and data products.
Selected work
- Kumah, K. K., Zandi, O., & Behrangi, A. (2025). Retrospective Mapping of Global Snow and Ice Cover Beyond the Satellite Observational Era. Earth and Space Science. DOI
- Kumah, K. K., Maathuis, B. H. P., Hoedjes, J. C. B., & Su, Z. (2022). Near real-time estimation of high spatiotemporal resolution rainfall from cloud-top properties of MSG and CML rainfall intensities. Atmospheric Research. DOI
- Kumah, K. K., and coauthors (2021). The MSG Technique. Remote Sensing. DOI
- Kumah, K. K., and coauthors (2021). Rain Area Detection in South-Western Kenya. Sensors. DOI
- Kumah, K. K., and coauthors (2020). Combining MWL and MSG SEVIRI Satellite Signals. Atmosphere.
- Open dataset. Global Snow and Ice Cover 1980-1987, University of Arizona Research Data Repository. Dataset DOI
Service
Service & Recognition
- Past peer-review contributions Completed manuscript reviews for journals including Journal of Hydrometeorology, Hydrology and Earth System Sciences, Earth System Science Data, Remote Sensing, and Journal of Geophysical Research: Atmospheres.
- Research community service External proposal review contribution for the Swiss National Science Foundation; Working Group 3 member, COST Action SmartSense.
Selected media and research features
- APS Physics Magazine Mobile Phone Signals Hold Real-Time Rainfall Information Feature article on the use of mobile-phone network signals for near-real-time rainfall monitoring, including discussion of Dr. Kumah's CML-satellite rainfall research.
- University of Twente / ITC ITC algorithm detects rainfall in rural Africa Institutional research feature on Dr. Kumah's machine-learning approach combining satellite observations and phone-signal rainfall information for rainfall estimation in data-sparse regions.
- TEMBO Africa A First for Ghana: Turning Mobile Phone Signals into Near-Real Time Rainfall Maps Project article on CML-based near-real-time rainfall mapping in Ghana using mobile-network signals, satellite products, station observations, and machine-learning methods.
Mission
Mission Statement
Precipitation is essential for weather, climate, water resources, agriculture, and hazard monitoring, but it remains difficult to measure over oceans, polar regions, complex terrain, and data-sparse regions. My work aims to improve the reliability, transparency, and usefulness of satellite precipitation and cryosphere records by combining physical understanding, machine learning, independent observations, and careful validation. The broader goal is to support better climate monitoring, hydrologic understanding, weather-risk assessment, and water-resource decision-making.
Contact
Collaborations in satellite precipitation, validation, and cryosphere applications.
University of Arizona
Tucson, Arizona
Tucson, Arizona
I welcome collaborations related to satellite precipitation retrieval, product validation, high-latitude precipitation, ocean precipitation, cryosphere applications, and rainfall monitoring in data-sparse regions.
- Email: kkumah@arizona.edu / kkumahkwabena@gmail.com
- Google Scholar: scholar profile
- ORCID: 0000-0001-6865-0665
- GitHub: king2coding
- LinkedIn: linkedin.com/in/kkumah