About

Jonathan L. Goodall is a Professor in the Department of Civil and Environmental Engineering at the University of Virginia and serves as the Director of the UVA Engineering Link Lab. His work focuses on water resources engineering, specifically advancing the field of hydroinformatics, where data and computational sciences are used to improve understanding, forecasting, and management of water systems. He collaborates with students in the Hydroinformatics Research Group to adapt techniques from artificial intelligence, machine learning, and cyber-physical systems for real-time flood mitigation and enhanced flood resiliency. Goodall is a registered Professional Engineer, a Fellow of the American Society of Civil Engineers, and an elected member of the Virginia Academy of Science, Engineering, and Medicine. His educational background includes a Ph.D. and M.S.E. in Civil Engineering from the University of Texas at Austin and a B.S. in Civil Engineering from the University of Virginia. His research interests encompass water resources engineering, hydroinformatics, stormwater flooding, smart cities, machine learning, and flood forecasting.

Research topics

• Computer Science
• Environmental science
• Engineering
• Geography
• Meteorology
• Artificial Intelligence
• Machine Learning
• Ecology
• Geotechnical engineering
• Political Science
• Geology
• Environmental planning
• Data science
• Cartography
• Systems engineering
• Business
• Telecommunications
• Water resource management
• Public relations
• Oceanography
• Distributed computing
• World Wide Web
• Transport engineering
• Civil engineering

Selected publications

• Experimental Evaluation of Serverless Data Layer Architectures for Smart City Internet of Things Applications

  Smart Cities · 2026-05-01

  articleOpen accessSenior author

  Comparative, experimentally grounded evidence for selecting smart city IoT data-layer architectures remains limited, complicating practical design decisions. This study provides an applied architecture decision-making guide by evaluating seven serverless data-layer architectures within a clearly defined service boundary (The Things Network, Azure-managed ingestion services, and Delta Lake persistence on object storage). Using a 21-day pilot deployment with nine LoRaWAN sensors, we compare ingestion completeness, median ingestion latency (estimated from TTN receive timestamps to Delta Lake commit times), cloud costs within an explicit boundary (ingestion, compute, and storage), and implementation/operational complexity proxies. Under the observed workload, TTN Storage Integration offers the lowest-cost archival ingestion via batching, Event Grid provides the most cost-effective near-real-time option among reliable pipelines, and Event Hubs demonstrates the highest ingestion completeness. The results are synthesized into practical guidance that maps common smart city application requirements to appropriate serverless ingestion patterns.

  PublisherDOI

• Comparing strategies for training LSTM models for street-scale urban flood prediction in Norfolk, Virginia

  Journal of Hydrology Regional Studies · 2026-04-18

  articleOpen access

  Norfolk, Virginia, United States For real-time urban flood prediction at the street-scale, both speed and accuracy are critical. Among deep learning algorithms, Long Short-Term Memory (LSTM) networks are effective for time series prediction. However, the optimal approach for training LSTM models for accurate prediction remains debated in the hydrology literature. Using a dataset of 40 flood-prone streets, this study explores strategies for training LSTM models for street-scale flood prediction. Three experiments were designed to (1) compare different data-grouping approaches across global, clustered, and street models, (2) assess how the availability of water-depth information influences prediction accuracy, and (3) test the scalability of the models for newly added streets. Grouping streets into hydrologically similar clusters enhanced prediction accuracy over the global model for the test events, while street models achieved the lowest errors in most cases. This suggests that the uniqueness of street-scale flooding dynamics in urban environments requires hyper-focused model training. When testing model performance with varying water-depth inputs, LSTM models trained on streets experiencing a wide range of flood depths performed well for streets with shallow water depths. The cluster models outperformed the global model in predicting flooding on newly added streets. This suggests that flooding behavior is better captured when the training dataset consists of hydrologically similar streets rather than a diverse set. • Strategies for training LSTM models for street-scale flood prediction are explored. • Training using only local data for a single street outperforms all other models. • Training using hydrologically similar streets outperforms training using all streets. • Training using streets experiencing deep flooding can still predict shallow flooding. • Urban flood dynamics require localized and specific training data.

  PublisherDOI

• Spatiotemporal Flood Forecasting in Coastal Urban Areas Using a Hybrid CNN-LSTM Surrogate Model

  SSRN Electronic Journal · 2025-01-01

  preprintOpen access
  PublisherDOI

• Deep learning-based downscaling of global digital elevation models for enhanced urban flood modeling

  Journal of Hydrology · 2025-01-16 · 25 citations

  article
  PublisherDOI

• Comparing Strategies for Training LSTM Models for Street-Scale Urban Flood Prediction

  SSRN Electronic Journal · 2025-01-01

  preprintOpen access
  PublisherDOI

• Evaluating Low Impact Development Practices as a Mitigation Strategy for Compound Flooding in Urban-Coastal Watersheds

  SSRN Electronic Journal · 2025-01-01

  preprintOpen accessSenior author
  PublisherDOI

• Integrating IoT Technology in Education: Enhancing Student Engagement and Local Flood Management in Coastal Virginia

  2025-04-22

  articleSenior author

  This work-in-progress paper details a project designed to engage Coastal Virginia middle-school students in climate resilience by deploying Internet of Things (IoT) sensors that collect real-time water quantity and quality data. By incorporating Dragino LoRaWAN technology and hands-on, classroom-friendly Arduino experiments, the project provides students with tangible skills in sensor technology and data analysis, connecting theoretical learning to real-world applications. Simultaneously, the data supports local efforts to improve regional hydrologic and hydraulic models.

  PublisherDOI

• Characterizing Stormwater Basin Conditions Using Tracked BMP Inspection and Rating Reports from the Virginia Department of Transportation

  Journal of Sustainable Water in the Built Environment · 2025-01-29 · 2 citations

  articleOpen accessSenior author

  Studies have rarely used stormwater best management practice (BMP) condition rating data to quantify changes in condition ratings and characterize condition issues, making it challenging to implement proactive maintenance practices. To help address this knowledge gap, we answer the following questions pertaining to a widely used stormwater BMP: detention and retention basins. (1) How often do basin condition ratings change over time? (2) What are specific site and structural condition issues identified during condition inspections? (3) What issues and site characteristics correspond with basins that experience a rapid decline in condition rating, meaning a decrease in two or more condition rating levels within a single year? We do this by characterizing basin condition ratings and associated issues using information included in more than 5,500 basin inspection reports, each containing more than 200 questions, stored in the Virginia Department of Transportation (VDOT) asset management system. On average, between 5.6 and 8.3 issues were recorded per visit for D-rated and E-rated basins compared with 0.03 to 1.8 issues for A-rated and B-rated basins. Of the 901 basins with three consecutive years of inspections, 41% and 35% had condition ratings that changed from 2020 to 2021 and from 2021 to 2022, respectively. The most common issues associated with rapid condition rating decline included major corrosion on the low flow orifice trash/debris rack, control structure, and pipe. Rapidly declining condition ratings were observed in 67% of VDOT’s nine independent management districts, suggesting rapid declines occur independently of management practices. Higher median elevation and percent slope attributes and lower median population were correlated with basins experiencing rapid condition rating decline. Using these results as context, we discuss considerations for improving BMP inspection data quality and opportunities for supporting proactive BMP management practices that can benefit other agencies managing a large portfolio of stormwater assets.

  PublisherOA PDFDOI

• Assessing the spatially heterogeneous transportation impacts of recurrent flooding in the Hampton roads region on auto accessibility

  Transportation Research Interdisciplinary Perspectives · 2025-11-01 · 1 citations

  articleOpen accessSenior author

  • Verified crowdsourced flood data reveals TAZ-level accessibility impacts. • Morning peak shows greatest losses: 1.7 % mean, up to 49.6 % for work access. • Lower education zones experience disproportionate flood accessibility reductions. • Event-day methodology captures spatial and temporal flooding heterogeneity. • Observed flood events replace simulations for measuring transport performance. Recurrent flooding has increased rapidly in coastal regions due to sea level rise and climate change. A key metric for evaluating transportation system degradation is accessibility, yet the lack of temporally and spatially disaggregate data means that the impact of recurrent flooding on accessibility—and hence transportation system performance—is not well understood. Using crowdsourced WAZE flood incident data from the Hampton Roads region in Virginia, this study examines changes in the roadway network accessibility for travelers residing in 1,113 traffic analysis zones (TAZs) across five time-of-day periods. Additionally, a social vulnerability index framework is developed to understand the socioeconomic characteristics of TAZs that experience high accessibility reduction under recurrent flooding. Results show that TAZs experience the most accessibility reduction under recurrent flooding during the morning peak period (6 to 9am) with large differences across different zones, ranging from 0 % to 49.6 % for work trips (with population-weighted mean reduction of 1.71 %) and 0 % to 87.9 % for non-work trips (with population-weighted mean reduction of 0.81 %). Furthermore, the social vulnerability analysis showed that zones with higher percentages of lower socio-economic status, unemployed, less educated, and limited English proficiency residents experience greater accessibility reduction for work trips. In contrast to previous studies that aggregate the effects of recurrent flooding across a city, these results demonstrate that there exists large spatial and temporal variation in recurrent flooding’s impacts on accessibility. This study also highlights the need to include social vulnerability analysis in assessing impacts of climate events, to ensure equitable outcomes as investments are made to create resilient transportation infrastructure.

  PublisherDOI

• Forecasting Multi-Step-Ahead Street-Scale Nuisance Flooding using a seq2seq LSTM Surrogate Model for Real-Time Application in a Coastal-Urban City

  Journal of Hydrology · 2025-01-13 · 22 citations

  articleOpen access

  In coastal-urban cities facing an elevated risk of nuisance flooding (by rain and tide) due to increased heavy rainfall, sea level rise, urbanization, and aging drainage systems, real-time flood forecasting at the street-scale can provide useful information to transportation decision-makers. Physics-Based Models (PBMs) that offer high accuracy come with high computational runtimes and costs that limit their application for real-time flood forecasting. To address this challenge, Machine Learning (ML) surrogate models trained from PBMs have been proposed to provide street-scale flood forecasts. Previous related studies have focused on using Long Short-Term Memory (LSTM) architectures to model hourly flood depth on streets. While LSTM models can capture input sequences effectively, they fall short in accurately preserving output sequences, limiting their suitability for multi-step-ahead forecasts. The seq2seq LSTM architecture offers a key advantage here by capturing the full sequence of input–output, making it potentially more suitable for multi-step-ahead flood forecasts compared to traditional LSTM models. However, seq2seq LSTM has not been tested for street-scale flood forecasting, particularly for rapidly fluctuating nuisance flooding events which require special attention to its temporal sequences. Hence, in this study, we applied the seq2seq LSTM model to explore multi-step-ahead street-scale nuisance flooding and compared its results to the traditional LSTM model as a benchmark model. LSTM and seq2seq LSTM surrogate models were applied to 22 flood-prone streets in Norfolk, Virginia, as a case study with a 4-hr (short-term) and 8-hr (long-term) lead time. The models were trained with environmental (rainfall and tide) and topographic (elevation, Topographic Wetness Index, and Depth-To-Water) features along with PBM-derived water depths for different storm events. The results demonstrated satisfactory performance of both LSTM and seq2seq LSTM surrogate models throughout the forecast period compared to the PBM. However, the seq2seq LSTM showed lower Mean Absolute Error (MAE)/ Root Mean Square Error (RMSE) and higher Nash–Sutcliffe Efficiency (NSE)/ correlation than the LSTM across most lead times, particularly for long-term forecasting due to its supremacy in handling both input–output sequences together, which is missing in the traditional LSTM. For example, in the long-term, the average RMSE ranges were 0.0268–0.0373 m for LSTM and 0.0226–0.0319 m for seq2seq LSTM, while in the short-term, they were 0.0263–0.0293 m and 0.0261–0.0283 m, respectively. Additionally, while both models exhibited similar performance in distinguishing flooded and non-flooded streets for flood depth ≥ 0.1 m, the seq2seq LSTM model demonstrated superior performance for higher flood depths (such as ≥ 0.2 m and ≥ 0.3 m). Once trained, inference took only 0.09 to 0.11 s (short-term) and 0.30 to 0.35 s (long-term) per storm event for the 22 streets, making the application highly suitable for real-time decision-making during nuisance flood events.

  PublisherDOI

Recent grants

• Collaborative Research: CiC (SEA): Using the Cloud to Model and Manage Large Watershed Systems

  NSF · $28k · 2014–2015

• NRT: A Graduate Traineeship in Cyber Physical Systems

  NSF · $3.6M · 2018–2024

• Workshop for Coordination of OpenMI and HIS Development in England, spring 2008

  NSF · $19k · 2008–2009

• CRISP Type 2: dMIST: Data-driven Management for Interdependent Stormwater and Transportation Systems

  NSF · $2.5M · 2017–2023

• Collaborative Research: CiC (SEA): Using the Cloud to Model and Manage Large Watershed Systems

  NSF · $185k · 2011–2014

Frequent coauthors

• Mohamed M. Morsy

  State Key Laboratory of Hydrology-Water Resources and Hydraulic Engineering

  114 shared

• Jeffrey M. Sadler

  Oklahoma State University

  92 shared

• Benjamin D. Bowes

  University of Virginia

  66 shared

• Anthony M. Castronova

  Consortium of Universities for the Advancement of Hydrologic Science

  56 shared

• Madhur Behl

  50 shared

• David G. Tarboton

  Utah State University

  50 shared

• Yawen Shen

  University of Virginia

  47 shared

• Faria Tuz Zahura

  Government of the United States of America

  40 shared

Labs

• Link LabPI

Awards & honors

• Fellow, American Society of Civil Engineers
• Early Career Research Excellence Award, International Enviro…
• CAREER Award, National Science Foundation (2009)
• Elected Member, Virginia Academy of Science, Engineering, an…