About

Dr. Catherine Nakalembe is an Assistant Professor of Geographical Sciences at the University of Maryland and the Director of the XylemLab. She also serves as the NASA Harvest Africa Program Lead. Her mission is to harness the power of satellite technology and artificial intelligence to support small-holder farmers and strengthen food systems across Africa. Dr. Nakalembe's work spans over 50 countries, where she delivers satellite-driven food security systems, disaster early warning tools, and AI programs that place advanced technology directly in the hands of Africa's farming communities. Her research group, the Xylem Lab, is an interdisciplinary Earth observation and AI research team focused on building satellite-based tools to monitor crops, detect food crises, and enhance climate resilience for farming communities in Africa and beyond. Dr. Nakalembe leads the NASA Harvest Africa Program, which has been active since 2016 across more than 15 countries. This flagship program uses Earth observation data to provide actionable food security intelligence for governments and humanitarian organizations continent-wide. Her innovative work includes developing helmet-mounted cameras combined with deep learning to generate high-resolution crop type datasets at scale, enabling AI-powered agricultural monitoring in regions where it was previously not possible. Her research also highlights the resilience of farmers during conflict, such as those in Tigray, Ethiopia, who continued cultivating crops amid civil war, challenging assumptions about war and food production. Her contributions have been recognized through numerous awards and honors, including being named a 2025 TED Fellow, and she has been featured in global media for her leadership in agricultural science and global food systems. Dr. Nakalembe collaborates extensively with governments, research institutions, humanitarian organizations, and media to bridge the gap between satellite science and the communities it serves, advancing scalable solutions for global challenges in food security and climate resilience.

Research topics

• Computer Science
• Geography
• Artificial Intelligence
• Environmental science
• Machine Learning
• Environmental resource management
• Agroforestry
• Forestry
• Telecommunications
• Engineering
• Business
• Agricultural engineering
• Ecology

Selected publications

• Optimizing satellite-based cropland area estimation through integrated map accuracy assessment and stratified sampling design across six African countries

  International Journal of Applied Earth Observation and Geoinformation · 2026-03-12

  articleOpen accessCorresponding

  Accurate cropland area estimation is essential for food security, yet conventional surveys are costly. Satellite-derived area estimates offer a scalable alternative; however, “pixel-counting” from satellite products introduces bias, while probability-based sampling with design-based inference provides unbiased area estimates, but its efficiency depends on map quality. In this study, we introduce an integrated framework linking map accuracy with relative efficiency to optimize stratified sampling designs. We evaluated seven land cover products across six African countries using independent reference data. Our results demonstrate that map selection is a critical determinant of survey costs: the most efficient products (predominantly GLAD and Digital Earth Africa) reduced required sample sizes by 20%–40% compared to less efficient alternatives while maintaining a target 10% coefficient of variation. We found that for rare cropland classes, gains in producers’ accuracy improved sampling efficiency more than equivalent gains in users’ accuracy. Our sample-based estimates ranged from 1.41 Mha (Rwanda) to 12.66 Mha (Tanzania) and correlated strongly with FAOSTAT arable land statistics ( R 2 = 0 . 89 ). Our framework minimizes the sample size required to achieve a target precision, providing an operational-ready, cost-effective, and statistically rigorous guide for national agricultural monitoring in resource- and data-limited regions. • We introduce an integrated framework linking map accuracy assessment with efficiency analysis to minimize the sample size required to achieve target precision. • Efficient maps reduce sample size requirements by 20%–40%, directly lowering reference data collection costs. • Producer’s accuracy gains improve efficiency more than user’s accuracy gains for rare cropland classes. • Sample-based area estimates achieved 8%–15% precision, outperforming conventional methods (27%–36%) across six African countries. • No single global product is optimal; country-specific evaluation is essential for cost-effective area estimation.

  PublisherDOI

• GeoAI-Based Crop Yield Estimation in Africa: A Systematic and Bibliometric Literature Review With Comparisons to Major Agricultural Producers

  SSRN Electronic Journal · 2026-01-01

  preprintOpen access
  PublisherDOI

• Climatic heterogeneity and constrained land access threaten the resilience of Uganda’s refugee agricultural self-reliance and integration model

  SSRN Electronic Journal · 2026-01-01

  preprintOpen access
  PublisherDOI

• Lessons From Uganda's Earth Observation‐Based Disaster Risk Financing Program

  AGU Advances · 2026-01-02 · 1 citations

  articleOpen access1st authorCorresponding

  Abstract Earth observation (EO) technologies are increasingly driving parametric insurance and risk financing for climate disasters, yet few operational programs demonstrate effective integration within national government systems. Uganda's Disaster Risk Financing Program (2016–2020) provides a rare example of satellite‐triggered financing operating at scale. Using MODIS vegetation indices to trigger drought response, the $14 million program supported over 452,000 people. It generated $11.1 million in immediate emergency aid savings, achieving a total return on investment of approximately 2.9 and an Internal Economic Rate of Return of 28.2%. This commentary synthesizes lessons from program implementation, highlighting that institutional and financial barriers, rather than technical limitations, now constrain the scaling of this EO‐driven climate resilience mechanism. While the program successfully integrated satellite data with transparent triggers and financial instruments, its sustainability depended on financial commitment extending beyond experimental phases. As climate risks intensify globally, Uganda's experience demonstrates that data‐triggered financing can operate within government institutions, but successful replication requires prioritizing institutional architecture and sustained financing over technical perfection.

  PublisherOA PDFDOI

• Uneven hydroclimatic risk and land access constraints challenge refugee agricultural self-reliance in Uganda

  Journal of Agriculture and Food Research · 2026-04-21

  articleOpen access

  Uganda's refugee policy is widely regarded for promoting agricultural self-reliance among the country’s nearly 2 million refugees, the largest refugee population in Africa. However, viability of this model is often challenged by how hydroclimatic shocks interact with household capacity to absorb them, an interaction that remain poorly understood. Here, we address a key gap by explicitly distinguishing unimodal and bimodal rainfall regimes and linking regime-specific drought and wetness dynamics to refugee livelihood exposure and household land access. We analyse 6-month Standardized Precipitation Evapotranspiration Index (1981-2024) and Vegetation Condition Index (2001-2024) anomalies, focusing on May-October main growing and harvest season. We find that the interannual extremes are highly episodic and often asynchronous between rainfall regimes. Only 2009, 2016, and 2023 exhibit widespread drought in both regimes, whereas only 2011 and 2012 are widely wet in both. Many large events are regime-specific, producing spatially uneven exposure, and long-term trends in affected area are weak relative to year-to-year variability. Major droughts are characterized by a late-season moisture deficit compounded by elevated atmospheric demand, with vegetation responses that are broadly consistent but spatially heterogeneous. We further show that hydroclimatic shocks intersect with rapidly rising refugee exposure in a small set of districts and with large structural constraints on land access and food-related outcomes. Notably, refugees typically hold one-eighth of the mean arable land as host households, a structural deficit suggesting that climate shocks rapidly and unevenly translate into production shortfalls rising market and aid dependence. These findings underscore the fragility of the self-reliance model under spatially uneven, episodic drought conditions. Sustaining refugee self-reliance will not only require spatially explicit climate monitoring but also need structural interventions in refugee-hosting regions where climatic, demographic, and land access pressures converge. • Uganda’s refugee-hosting districts span distinct unimodal and bimodal rainfall regimes. • Widespread drought in both regimes occurred only in 2009, 2016, and 2023. • Bimodal areas show more moderate droughts, while severe unimodal events last longer. • Refugees typically hold one-eighth as much arable land as host households. • Climate shocks can rapidly and unevenly translate into production shortfalls disproportionately affecting refugees.

  PublisherDOI

• EO-based long term cropland and paddy monitoring with the farm action toolkit (FAcT): strengthening policy support in Bhutan

  Discover Agriculture · 2026-01-29

  articleOpen access

  , an AI-enabled, Earth Observation (EO)-based framework for long-term cropland and paddy monitoring (2002-2024), linking EO data to farmer benefit access and FYP implementation. FAcT delivers Bhutan's first national, field-scale cropland and paddy dataset, achieving 87-92% accuracy and [Formula: see text] values of 0.75--0.85 against government statistics. Between 2002 and 2024, cropland experienced a 22.5% net increase, with 50.3% gain and 27.8% loss with respect to the 2002 baseline. Of the cropland lost, 66% reverted to forest post-2018, aligning with 12th FYP conservation goals. Net primary productivity declined by 2%, while per-capita cropland area dropped by 16.5%, underscoring population pressure and land competition. In Paro District, approximately 30% of cultivated land verified using EO data was found to be active but missed policy benefits due to gaps in manual verification, revealing critical inclusion barriers. FAcT's co-development with national agencies ensured scientific rigor and institutional uptake. The open-source toolkit (https://zenodo.org/records/15621464) supports land-use decision-making, resilience planning, and sustainable agriculture in smallholder, high-elevation, and data-scarce contexts. Findings contribute to Sustainable Development Goals (SDGs) 2, 11, and 15, demonstrating how EO-based agricultural monitoring can inform policy interventions and impact tracking in mountainous regions. Supplementary Information: The online version contains supplementary material available at 10.1007/s44279-026-00498-3.

  PublisherOA PDFDOI

• Assessing maize and cassava extent and intercropping in southwest Nigeria

  Environmental Research Food Systems · 2026-03-31

  articleOpen access

  Abstract Smallholder systems contribute to nearly three-quarters of food production in sub-Saharan Africa and employ half of the continent’s population. At the same time, these primarily rainfed systems are sensitive to climate variability and have limited capital for obtaining buffering inputs. Both their significance and vulnerability require improved agricultural monitoring. However, the small field sizes of smallholder farms (<2 ha), their complex multi- and intercropped systems, and persisting growing season cloud cover in tropical and sub-tropical regions where many smallholder systems occur mean that the fine-scale cropping patterns of smallholder landscapes remain poorly understood. Focusing on the Oyo state in southwest Nigeria, the country with the largest smallholder population in Africa, we seek to overcome these challenges by mapping the extent of maize and cassava cultivation, two of the country’s most widely cultivated staple crops. Using geolocated GoPro images collected as ground-truth data, we trained machine/deep-learning models (i.e. random forest and convolutional neural network) to classify high-resolution (10 m) multispectral satellite imagery. In doing so, we estimated that 175 604 ha of cassava and 150 538 ha of maize would be cultivated across the state in 2022, in close agreement with official statistics and coarser global crop-type products. We then evaluated a suite of dry-season vegetation indices to assess the prevalence of maize–cassava intercropping (in which maize is harvested at the end of the rainy season and cassava persists), finding indicative evidence of the practice in more than a third (45.2%) of maize locations. Thus, our findings challenge conventional remote sensing assumptions (i.e. monocultured cultivation) and provide important advances for scalable and transferable fine-scale crop type mapping in complex smallholder systems.

  PublisherDOI

• Multi-temporal Assessment of Flood Damage in the Sub-Lower Niger River Basin, Nigeria with Multi-sensors and Google Earth Engine

  2026-01-01

  book-chapter
  PublisherDOI

• Maps of Cassava and Maize Extent in Oyo State, Nigeria (year 2022)

  Zenodo (CERN European Organization for Nuclear Research) · 2025-10-30

  otherOpen access

  This is a geospatial raster dataset (10 m resolution) mapping the spatial distribution of maize and cassava fields in Oyo State, southwestern Nigeria, for the year 2022. Crop predictions were derived using fused Sentinel-1 SAR and Sentinel-2 optical imagery combined with supervised machine-learning classification and field-validated ground-truth observations. The dataset supports applications in food security monitoring, land-use analysis, and agricultural management in smallholder systems.

  PublisherDOI

• A framework for EO-based National Agricultural Monitoring (EO-NAM) for the African Context

  npj Sustainable Agriculture · 2025-08-04

  letterOpen access1st authorCorresponding

  Effective agriculture monitoring is vital for food security and achieving UN Sustainable Development Goal 2: Zero Hunger. Earth-observations (EO) offer unparalleled potential for scalable data, yet many developing nations, particularly in Africa, face challenges due to limited investments in human capacity and technology. We present a phased framework for EO-based agriculture monitoring systems, emphasizing national commitment and leveraging existing structures for long-term sustainability and adopting and adapting future advancements.

  PublisherOA PDFDOI

Frequent coauthors

• Inbal Becker‐Reshef

  University of Maryland, College Park

  39 shared

• Hannah Kerner

  35 shared

• Gabriel Tseng

  McGill University

  24 shared

• Christina Justice

  University of Maryland, College Park

  16 shared

• Ivan Zvonkov

  University of Maryland, College Park

  15 shared

• B. Barker

  University of Maryland, College Park

  12 shared

• M. L. Humber

  University of Maryland, College Park

  9 shared

• Jan Dempewolf

  Bavarian State Research Center for Agriculture

  9 shared

Education

• Ph.D., Geography

  University of Maryland

  2007

• M.S., Geography

  University of Maryland

  2003

• B.A., Geography

  University of Maryland

  2001

Awards & honors

• 2020 Africa Food Prize Laureate
• 2020 UMD Research Excellence Honoree
• Inaugural GEO Individual Excellence Award (2019)