Search NASA⌕ Search

Engineering topics

Christopher R Hain

Publications and source records attributed to Christopher R Hain.

13 records
All sourcesNASA NTRSDOE OSTI
NASA NTRS2026

Hydrology Copilot: A Cloud-Native Ai System for Hydrological Data Analysis

The emergence of AI-driven Earth observation systems promises to broaden access to petabyte-scale geospatial data beyond domain specialists. However, translating this vision into operational scientific infrastructure requires addressing fundamental challenges in data virtualization, code transparency, and domain-specific reasoning. We present Hydrology Copilot, a cloud-native AI framework for natural-language-driven analysis of Earth observation data. To demonstrate operational capabilities at scale, we implement the system using NASA's North American Land Data Assimilation System version 3 (NLDAS-3), which provides surface meteorological forcing and land-surface model output across North and Central America at 1-km resolution, from which drought diagnostics are derived. The system integrates five core contributions: (1) scalable data virtualization using Kerchunk-based cloud optimized access, achieving a 1.5 to 4.6 times improvement in I/O latency across benchmark queries spanning regional single-day extractions (4.6 times speedup) to continental monthly aggregations (1.5 times speedup); (2) transparent code generation through Microsoft Azure AI Foundry agents that expose executable Python workflows for scientific verification; (3) persistent conversational memory enabling multi-turn analytical discourse across sessions; (4) intelligent query validation that enforces dataset boundaries and resolves ambiguous requests before execution; and (5) a multi-agent architecture coordinating query parsing, code generation, and visualization. We evaluate the system through drought-monitoring workflows, demonstrating reliable code generation, accurate results validated against reference computations and the operational U.S. Drought Monitor, and efficient operation across increasingly complex tasks. By bridging natural-language interfaces with rigorous hydrological analysis, Hydrology Copilot advances beyond proof-of-concept demonstrations to provide a deployable framework for operational Earth science applications.

Data virtualization↗
NASA NTRS2024

Prithvi-EO-2.0: A Versatile Multi-Temporal Foundation Model for Earth Observation Applications

This technical report presents Prithvi-EO-2.0, a new geospatial foundation model that offers significant improvements over its predecessor, Prithvi-EO-1.0. Trained on 4.2M global time series samples from NASA’s Harmonized Landsat and Sentinel-2 data archive at 30m resolution, the new 300M and 600M parameter models incorporate temporal and location embeddings for enhanced performance across various geospatial tasks. Through extensive benchmarking with GEOBench, the 600M version outperforms the previous Prithvi-EO model by 8% across a range of tasks. It also outperforms six other geospatial foundation models when benchmarked on remote sensing tasks from different domains and resolutions (i.e. from 0.1m to 15m). The results demonstrate the versatility of the model in both classical earth observation and high-resolution applications. Early involvement of end-users and subject matter experts (SMEs) are among the key factors that contributed to the project’s success. In particular, SME involvement allowed for constant feedback on model and dataset design, as well as successful customization for diverse SME-led applications in disaster response, land use and crop mapping, and ecosystem dynamics monitoring. Prithvi-EO-2.0 is available on Hugging Face and IBM terratorch, with additional resources on GitHub. The project exemplifies the Trusted Open Science approach embraced by all involved organizations.

Daniela Szwarcman↗
NASA NTRS2024

Assessing Flash Characteristics in Lightning-Initiated Wildfire Events Between 1995 and 2020 Within the Contiguous United States

Twenty-six years of lightning data were paired with over 68,000 lightning-initiated wildfire (LIW) reports to understand lightning flash characteristics responsible for ignition in between 1995 and 2020. Results indicate that 92% of LIW were started by negative cloud-to-ground (CG) lightning flashes and 57% were single stroke flashes. Moreover, 62% of LIW reports did not have a positive CG within 10 km of the start location, contrary to the science literature’s suggestion that positive CG flashes are a dominant fire-starting mechanism. Nearly 1/3rd of wildfire events were holdovers, meaning one or more days elapsed between lightning occurrence and fire report. However, fires that were reported less than a day after lightning occurrence statistically burned more acreage. Peak current was not found to be a statistically significant delineator between fire starters and non-fire starters for -CGs but was for positive CGs. Results highlighted the need for reassessing the role of positive CG lightning and subsequently long continuing current in wildfire ignition started by lightning. One potential outcome of this study’s results is the development of real-time tools to identify ignition potential during lightning events to aid in fire mitigation efforts.

Christopher J Schultz↗
NASA NTRS2022

Assimilation of GPM-retrieved Ocean Surface Meteorology Data for Two Snowstorm Events during ICE-POP 2018

As a component of the National Aeronautics and Space Administration (NASA) Weather Focus Area and Global Precipitation Measurement (GPM) Ground Validation participation in the International Collaborative Experiments for PyeongChang 2018 Olympic and Paralympic Winter Games (ICE-POP 2018) field research and forecast demonstration programs, hourly ocean surface meteorology properties were retrieved from the GPM microwave observations for January – March 2018. In this study, the retrieved ocean surface meteorological products – 2-m temperature, 2-m specific humidity, and 10-m wind speed were assimilated into a regional numerical weather prediction (NWP) framework to explore the application of these observations for two heavy snowfall events during the ICE-POP 2018: 27-28 February, and 7-8 March 2018. The Weather Research and Forecasting (WRF) model and the community Gridpoint Statistical Interpolation (GSI) were used to conduct high resolution simulations and data assimilation experiments. The results indicate that the data assimilation has a large influence on surface thermodynamic and wind fields in the model initial condition for both events. With cycled data assimilation, significantly positive influence of the retrieved surface observation was found for the March case with improved quantitative precipitation forecast and reduced error in temperature forecast. A slightly smaller yet positive impact was also found in the forecast of the February case.

assimilation↗
NASA NTRS2021

Sustainable Use of Groundwater May Dramatically Reduce Irrigated Production of Maize, Soybean, and Wheat

Groundwater extraction in the United States (US) is unsustainable, making it essential to understand the impacts of limited water use on irrigated agriculture. To improve this understanding, we integrated a gridded crop model with satellite observations, recharge estimates, and water survey data to assess the effects of sustainable groundwater withdrawals on US irrigated agricultural production. The gridded crop model agrees with satellite-based estimates of evapotranspiration (R2 = 0.68), as well as survey data from the United States Department of Agriculture (R2 = 0.82–0.94 for county-level production and 0.37–0.54 for county-level yield). Using the optimistic assumption that groundwater extraction equals effective aquifer recharge rate, we find that sustainable groundwater use decreases US irrigated production of maize, soybean, and winter wheat by 20%, 6%, and 25%, respectively. Using a more conservative assumption of groundwater availability, US irrigated production of maize, soybean, and winter wheat decreases by 45%, 37%, and 36%, respectively. The wide range of simulated losses is driven by considerable uncertainty in surface water and groundwater interactions, as well as accounting for the many aspects of sustainability. Our results demonstrate the vulnerability of US irrigated agriculture to unsustainable groundwater pumping, highlighting the difficulty of expanding or even maintaining irrigated food production in the face of climate change, population growth, and shifting dietary demands. These findings are based on reducing pumping by fallowing irrigated farmland; however, alternate pumping reduction strategies or technological advances in crop genetics and irrigation could produce different results.

sustainability↗
NASA NTRS2018

An Initial Assessment of a SMAP Soil Moisture Disaggregation Scheme Using TIR Surface Evaporation Data over the Continental United States

The Soil Moisture Active Passive (SMAP) mission is dedicated toward global soil moisture mapping. Typically, an L-band microwave radiometer has spatial resolution on the order of 36-40 km, which is too coarse for many specific hydro-meteorological and agricultural applications. With the failure of the SMAP active radar within three months of becoming operational, an intermediate (9-km) and finer (3-km) scale soil moisture product solely from the SMAP mission is no longer possible. Therefore, the focus of this study is a disaggregation of the 36-km resolution SMAP passive-only surface soil moisture (SSM) using the Soil Evaporative Efficiency (SEE) approach to spatial scales of 3-km and 9-km. The SEE was computed using thermal-infrared (TIR) estimation of surface evaporation over Continental U.S. (CONUS). The disaggregation results were compared with the 3 months of SMAP-Active (SMAP-A) and Active/Passive (AP) products, while comparisons with SMAP-Enhanced (SMAP-E), SMAP-Passive (SMAP-P), as well as with more than 180 Soil Climate Analysis Network (SCAN) stations across CONUS were performed for a 19 month period. At the 9-km spatial scale, the TIR-Downscaled data correlated strongly with the SMAP-E SSM both spatially (r = 0.90) and temporally (r = 0.87). In comparison with SCAN observations, overall correlations of 0.49 and 0.47; bias of −0.022 and −0.019 and unbiased RMSD of 0.105 and 0.100 were found for SMAP-E and TIR-Downscaled SSM across the Continental U.S., respectively. At 3-km scale, TIR-Downscaled and SMAP-A had a mean temporal correlation of only 0.27. In terms of gain statistics, the highest percentage of SCAN sites with positive gains (>55%) was observed with the TIR-Downscaled SSM at 9-km. Overall, the TIR-based downscaled SSM showed strong correspondence with SMAP-E; compared to SCAN, and overall both SMAP-E and TIR-Downscaled performed similarly, however, gain statistics show that TIR-Downscaled SSM slightly outperformed SMAP-E.

Disaggregation↗
NASA NTRS2017

The Role of Local Heating in the 2015 Indian Heat Wave

India faced a major heat wave during the summer of 2015. Temperature anomalies peaked in the dry period before the onset of the summer monsoon, suggesting that local land-atmosphere feedbacks involving desiccated soils and vegetation might have played a role in driving the heat extreme. Upon examination of in situ data, reanalysis, satellite observations, and land surface models, we find that the heat wave included two distinct peaks: one in late May, and a second in early June. During the first peak we find that clear skies led to a positive net radiation anomaly at the surface, but there is no significant sensible heat flux anomaly within the core of the heat wave affected region. By the time of the second peak, however, soil moisture had dropped to anomalously low levels in the core heat wave region, net surface radiation was anomalously high, and a significant positive sensible heat flux anomaly developed. This led to a substantial local forcing on air temperature that contributed to the intensity of the event. The analysis indicates that the highly agricultural landscape of North and Central India can reinforce heat extremes under dry conditions.

Debjani Ghatak↗
NASA NTRSDate not supplied

A SATELLITE AGNOSTIC APPROACH TO QUANTIFYING HAIL DAMAGE SWATHS ACROSS THE CENTRAL UNITED STATES AND OTHER AGRICULTURAL REGIONS

Intense thunderstorms can bring damaging winds and large hail to agricultural regions during the prime growing season. In certain cases, large swaths of damage from the wind and hail are left behind and visible to satellite remote sensing instruments. Often times, Earth observing optical remote sensing from low to high spatial resolutions are able to view these damaged swaths. With the large number of moderate to high-resolution instruments in orbit, these damaged areas have potential to be viewed daily. However, during the prime growing season, clouds frequently block the viewing of the land surface by these optical instruments. Space-borne synthetic aperture radar (SAR) instruments allow for the viewing of the land surface in most weather conditions, but instead measure backscatter as opposed to optical sensors measuring reflected or emitted radiation. Additionally, the number of SAR instruments with free and open data lags behind the number of optical sensors. This paper describes the development of a methodology that attempts to characterize hail damaged swaths, through independent use of multiple optical and SAR platforms. This satellite-agnostic approach will focus statistical analysis by comparing undamaged areas to suspected damaged areas by using commonly derived indices from optical instruments and SAR backscatter from multiple polarizations.

Jordan R Bell↗
NASA NTRSDate not supplied

Role of Antecedent Soil Moisture and Vegetation Stress in Lightning-Initiated Wildfires

Lightning-caused wildfires are a small percentage of all wildfire events within the Conterminous U.S. (CONUS), but they account for over 56% of the acreage burned. The atmospheric conditions favoring wildfire and rapid growth are well understood: large dewpoint depressions, unstable planetary boundary layer, strong winds, etc. However, antecedent land surface conditions affecting dead and live fuel moisture is more difficult to quantify. This study examines over 20 years of antecedent land surface, vegetation stress, and wildfire characteristic data associated with nearly 77,000 lightning-initiated wildfires from the U.S. Forest Service Wildfire Database. We will invoke two in-house databases generated by the NASA Short-term Prediction Research and Transition (SPoRT) Center: an observations-driven, climatological run of the Noah land surface model within the NASA Land Information System (i.e., SPoRT-LIS) to depict soil moisture deficits / anomalies, and a satellite-constrained Evaporative Stress Index (ESI) product to denote areas of stressed vegetation. We will mine these datasets associated with lightning-caused (and null) events to determine important relationships, distributions, and delineators that correspond to elevated threat areas for lightning-initiated wildfires.

Wildfire↗
NASA NTRSDate not supplied

Predicting Lightning Initiation using Deep Learning

Lightning occurrence presents safety challenges to people and property. The main challenge with lightning safety is that the majority of guidance is reactive. In other words, lightning has to have already occurred nearby before a person will respond and take shelter. Further, most injuries or fatalities occur as the storm approaches, or as it's moving away, when rainfall may not be present at the time of the flash. Thus, this project develops a physically-based deep learning model to produce lightning probabilities out to 15 minutes. The deep learning model combines a Convolutional Neural Network (CNN) with a Long Short-Term Memory (LSTM) network to capture both the spatial and temporal evolution of storms to predict the probability that lightning initiation will occur in the next 15 minutes. The model combines radar reflectivity, correlation coefficient and differential reflectivity to inferred storm hydrometer type and precipitation phase, which aids in the identification of electrification processes. The model is trained with data from the Geostationary Lightning Mapper (GLM), which is a near infrared sensor onboard the GOES-R series of satellites that measures optical brightness from lightning. This presentation will provide an overview of the project.

Andrew T White↗
NASA NTRSDate not supplied

Urban Expansion and Climate Change: Investigating the Impact of Future Scenarios on Child Health

Cities in the poorest countries of the world are rapidly urbanizing. In fact, some of the fastest growing cities on the planet are found in sub-Saharan Africa. Many people benefit from economic, educational, and health opportunities that exist in cities and, as a result, city-dwellers often face less risk of poverty and disease in the context of climate change than their rural-dwelling counterparts. However, rapid urbanization, resulting from a large and constant influx of migrants from rural areas, can put pressure on existing social, economic, and health systems. For poor cities, they face a challenge to support existing residents while also expanding to support the needs of migrants, many of whom are deeply impoverished. Additionally, in a climate change, linked to heat waves, floods, droughts, and other related events, can add stress to complex urban food, health, hygiene, and housing systems. In this project we investigate urban expansion, climate change, and health in four major African cities – Ouagadougou, Addis Ababa, Nairobi, and Kigali. We consider different scenarios of urban expansion using historic and contemporary maps of urban extent, combined with climate (temperature, rainfall, and vegetation) and land cover combined with spatially referenced health information from the Demographic and Health Surveys (DHS) to investigate the relationships between urban dwelling, climate change and health. Temperature and rainfall scenarios are developed under different urban land use futures (expanding agricultural areas versus reducing agricultural areas) to examine the ways that individual health outcomes related to malnutrition vary under different potential future conditions. Preliminary results of the research highlight the importance of temperature for health, in particular, and suggest that while urban conditions related to urban infrastructure (e.g., educational attainment, electricity access, and improved hygiene) cannot reduce health risks to counter the impacts of high temperatures and reduced agricultural land. As cities in poor countries urbanize, city-dwellers face unique risks when droughts, heat waves, and other extreme weather events occur. In this project we explore urban land use conditions, climate conditions and health using future scenarios. Specifically we evaluate health outcomes in the future while considering different land use practices, urban expansion, temperature, and rainfall conditions to identify individual-level risk and protective factors. We compare results across four major cities in sub Saharan Africa.

Public Health↗
NASA NTRSDate not supplied

NASA SPoRT’s Streamflow-AI: Updates and Advancements

The NASA Short-term Prediction Research and Transition Center (SPoRT) has been running a near-real-time deep learning model, Streamflow-AI, that predicts stream heights at over 250 locations across the Eastern United States out to 7 days for several years. This project was born out of a research-to-operations/operation-to-research (R2O/O2R) paradigm within SPoRT through extensive collaboration with National Weather Service (NWS) and River Forecast Center (RFC) forecasters. The original intention of Streamflow-AI was to provide accurate medium range (>2 days) forecasts of river flooding events that are particularly challenging for NWS forecasters, because routine streamflow forecasts may not incorporate quantitative precipitation forecast (QPF) data beyond one or two days. Therefore, Streamflow-AI was designed to help fill the operational gap, producing routine 7 day river level forecasts using a suite of QPFs. Throughout the evolution of the project, research efforts have been tailored to meet the needs of the end users. This presentation will provide a brief overview of NASA SPoRT’s Streamflow-AI product, highlighting O2R/R2O concepts with a focus on recent developments including a 1-hr rapid update model and 15-min burn scar flooding modeling efforts.

Flooding↗