David Sathiaraj
Research Subjects
David Sathiaraj has a total of 29 co-authors. Here are the top co-authors by number of shared publications:
- Barry D. Keim
Department of Geography & Anthropology, Louisiana State University, Baton Rouge, LA 70803, USA
4 shared publications - Hal F. Needham
Louisiana State University
3 shared publications Barry D. Keim
Department of Geography and Anthropology Louisiana State University Baton Rouge Louisiana USA
2 shared publicationsEric Rohli
AI and Data Science, Trabus Technologies, San Diego, CA, USA
2 shared publications- Mark A. Shafer
Louisiana State University, Department of Geography and Anthropology
2 shared publications - Aubrey Dugger
National Science Foundation National Center for Atmospheric Research Boulder CO USA
1 shared publication
Latest Publications
- Open Access
- 25 July 2023
- Published by Cambridge University Press (CUP) in Environmental Data Science
- Vol. 2, e30
Near-term forecasting of water reservoir storage capacities using long short-term memory
Predicting reservoir storage capacities is an important planning activity for effective conservation and water release practices. Weather events such as drought and precipitation impact water storage capacities in reservoirs. Predictive insights on reservoir storage levels are beneficial for water planners and stakeholders in effective water resource management. A deep learning (DL) neural network (NN) based reservoir storage prediction approach is proposed that learns from climate, hydrological, and storage information within the reservoir’s associated watershed. These DL models are trained and evaluated for 17 reservoirs in Texas, USA. Using the trained models, reservoir storage predictions were validated with a test data set spanning 2 years. The reported results show promise for longer-term water planning decisions.
- 25 July 2023
- Published by Cambridge University Press (CUP) in Environmental Data Science
- Vol. 2, e30
Open AccessNear-term forecasting of water reservoir storage capacities using long short-term memory
Predicting reservoir storage capacities is an important planning activity for effective conservation and water release practices. Weather events such as drought and precipitation impact water storage capacities in reservoirs. Predictive insights on reservoir storage levels are beneficial for water planners and stakeholders in effective water resource management. A deep learning (DL) neural network (NN) based reservoir storage prediction approach is proposed that learns from climate, hydrological, and storage information within the reservoir’s associated watershed. These DL models are trained and evaluated for 17 reservoirs in Texas, USA. Using the trained models, reservoir storage predictions were validated with a test data set spanning 2 years. The reported results show promise for longer-term water planning decisions.
- 1 November 2018
- Published by Elsevier in Computers, Environment and Urban Systems
- Vol. 72, 212-220
Data-driven analysis on the effects of extreme weather elements on traffic volume in Atlanta, GA, USA
- 1 November 2018
- Published by Elsevier in Computers, Environment and Urban Systems
- Vol. 72, 212-220
Data-driven analysis on the effects of extreme weather elements on traffic volume in Atlanta, GA, USA
- 14 August 2018
- Published by Wiley in Environmetrics
- Vol. 30 (4)
Predicting climate types for the Continental United States using unsupervised clustering techniques
The problem of clustering climate data observation sites and grouping them by their climate types is considered. Machine learning–based clustering algorithms are used in analyzing climate data time series from more than 3,000 climate observation sites in the United States, with the objective of classifying climate type for regions across the United States. Understanding the climate type of a region has applications in public health, environment, actuarial science, insurance, agriculture, and engineering.In this study, daily climate data measurements for temperature and precipitation from the time period 1946–2015 have been used. The daily data observations were grouped into three derived data sets: a monthly data set (daily data aggregated by month), an annual data set (daily data aggregated by year), and a threshold exceeding frequency data set (threshold exceeding frequency provides the frequency of occurrence of certain climate extremes). Three existing clustering algorithms from the literature, namely, k‐means, density‐based spatial clustering of applications with noise, and balanced iterative reducing and clustering using hierarchies, were each applied to cluster each of the data sets, and the resulting clusters were assessed using standardized clustering indices. The results from these unsupervised learning techniques revealed the suitability and applicability of these algorithms in the climate domain. The clusters identified by these techniques were also compared with existing climate classification types such as the Köppen classification system. Additionally, the work also developed an interactive web and map‐based data visualization system that uses efficient big data management techniques to provide clustering solutions in real time and to display the results of the clustering analysis.
- 14 August 2018
- Published by Wiley in Environmetrics
- Vol. 30 (4)
Predicting climate types for the Continental United States using unsupervised clustering techniques
The problem of clustering climate data observation sites and grouping them by their climate types is considered. Machine learning–based clustering algorithms are used in analyzing climate data time series from more than 3,000 climate observation sites in the United States, with the objective of classifying climate type for regions across the United States. Understanding the climate type of a region has applications in public health, environment, actuarial science, insurance, agriculture, and engineering.In this study, daily climate data measurements for temperature and precipitation from the time period 1946–2015 have been used. The daily data observations were grouped into three derived data sets: a monthly data set (daily data aggregated by month), an annual data set (daily data aggregated by year), and a threshold exceeding frequency data set (threshold exceeding frequency provides the frequency of occurrence of certain climate extremes). Three existing clustering algorithms from the literature, namely, k‐means, density‐based spatial clustering of applications with noise, and balanced iterative reducing and clustering using hierarchies, were each applied to cluster each of the data sets, and the resulting clusters were assessed using standardized clustering indices. The results from these unsupervised learning techniques revealed the suitability and applicability of these algorithms in the climate domain. The clusters identified by these techniques were also compared with existing climate classification types such as the Köppen classification system. Additionally, the work also developed an interactive web and map‐based data visualization system that uses efficient big data management techniques to provide clustering solutions in real time and to display the results of the clustering analysis.
- Open Access
Modeling Hydroclimatic Change in Southwest Louisiana Rivers
We applied the newly developed WRF-Hydro model to investigate the hydroclimatic trend encompassing the three basins in Southwest Louisiana as well as their connection with large-scale atmospheric drivers. Using the North American Land Data Assimilation System Phase 2 (NLDAS-2), we performed a multi-decadal model hindcast covering the period of 1979–2014. After validating the model’s performance against available observations, trend and wavelet analysis were applied on the time series of hydroclimatic variables from NLDAS-2 (temperature and precipitation) and model results (evapotranspiration, soil moisture, water surplus, and streamflow). Trend analysis of model-simulated monthly and annual time series indicates that the regional climate is warming and drying over the past decades, specifically during spring and summer (growing season). Wavelet analysis reveals that, since the late 1990s, the anomaly of evapotranspiration, soil moisture, and streamflow exhibits high coherency with that of precipitation. Pettitt’s test detects a possible change-point around the year 2004, after which the monthly precipitation decreased from 140 to 120 mm, evapotranspiration slightly increased from 80 to 83 mm, and water surplus decreased from 60 to 38 mm. Changes in regional climate conditions are closely correlated with large-scale climate dynamics such as the Atlantic Multidecadal Oscillation (AMO) and El Niño Southern Oscillation (ENSO).
Open AccessModeling Hydroclimatic Change in Southwest Louisiana Rivers
We applied the newly developed WRF-Hydro model to investigate the hydroclimatic trend encompassing the three basins in Southwest Louisiana as well as their connection with large-scale atmospheric drivers. Using the North American Land Data Assimilation System Phase 2 (NLDAS-2), we performed a multi-decadal model hindcast covering the period of 1979–2014. After validating the model’s performance against available observations, trend and wavelet analysis were applied on the time series of hydroclimatic variables from NLDAS-2 (temperature and precipitation) and model results (evapotranspiration, soil moisture, water surplus, and streamflow). Trend analysis of model-simulated monthly and annual time series indicates that the regional climate is warming and drying over the past decades, specifically during spring and summer (growing season). Wavelet analysis reveals that, since the late 1990s, the anomaly of evapotranspiration, soil moisture, and streamflow exhibits high coherency with that of precipitation. Pettitt’s test detects a possible change-point around the year 2004, after which the monthly precipitation decreased from 140 to 120 mm, evapotranspiration slightly increased from 80 to 83 mm, and water surplus decreased from 60 to 38 mm. Changes in regional climate conditions are closely correlated with large-scale climate dynamics such as the Atlantic Multidecadal Oscillation (AMO) and El Niño Southern Oscillation (ENSO).
- 1 December 2017
- Published by SAGE Publications in Big Data
- Vol. 5 (4) , 325-336
Improving Predictive Accuracy in Elections
The problem of accurately predicting vote counts in elections is considered in this article. Typically, small-sample polls are used to estimate or predict election outcomes. In this study, a machine-learning hybrid approach is proposed. This approach utilizes multiple sets of static data sources, such as voter registration data, and dynamic data sources, such as polls and donor data, to develop individualized voter scores for each member of the population. These voter scores are used to estimate expected vote counts under different turnout scenarios. The proposed technique has been tested with data collected during U.S. Senate and Louisiana gubernatorial elections. The predicted results (expected vote counts, predicted several days before the actual election) were accurate within 1%.
- 1 December 2017
- Published by SAGE Publications in Big Data
- Vol. 5 (4) , 325-336
Improving Predictive Accuracy in Elections
The problem of accurately predicting vote counts in elections is considered in this article. Typically, small-sample polls are used to estimate or predict election outcomes. In this study, a machine-learning hybrid approach is proposed. This approach utilizes multiple sets of static data sources, such as voter registration data, and dynamic data sources, such as polls and donor data, to develop individualized voter scores for each member of the population. These voter scores are used to estimate expected vote counts under different turnout scenarios. The proposed technique has been tested with data collected during U.S. Senate and Louisiana gubernatorial elections. The predicted results (expected vote counts, predicted several days before the actual election) were accurate within 1%.
- 1 November 2017
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 303-312
Deriving Data-Driven Insights from Climate Extreme Indices for the Continental US
Daily climate data observations from more than 3000 climate measurement sites in the continental U.S. were mined and analyzed to derive insights and trends from climate extreme indices. Daily climate data observations were aggregated by climate divisions and analyzed to derive a new climate extremes indices data set (Threshold Exceedence Frequency, TEF). Each climate division was statistically assessed for the following elements: maximum and minimum temperature, precipitation and snowfall. The climate data time series were divided into 2 time intervals (1946-1980 and 1981-2015) and the occurrence frequencies of various climate extreme indices was statistically examined. Results revealed interesting insights such as an increasing frequency of occurrence of night-time temperatures in South-east US and decreasing frequency of winter temperature and snowfall extremes in northern US. The study also produced a new web-based visualization system to analyze the results of the study. The visualization system included interactive choropleth maps and charts to depict spatiotemporal changes in various climate thresholds over time.
- 1 November 2017
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 303-312
Deriving Data-Driven Insights from Climate Extreme Indices for the Continental US
Daily climate data observations from more than 3000 climate measurement sites in the continental U.S. were mined and analyzed to derive insights and trends from climate extreme indices. Daily climate data observations were aggregated by climate divisions and analyzed to derive a new climate extremes indices data set (Threshold Exceedence Frequency, TEF). Each climate division was statistically assessed for the following elements: maximum and minimum temperature, precipitation and snowfall. The climate data time series were divided into 2 time intervals (1946-1980 and 1981-2015) and the occurrence frequencies of various climate extreme indices was statistically examined. Results revealed interesting insights such as an increasing frequency of occurrence of night-time temperatures in South-east US and decreasing frequency of winter temperature and snowfall extremes in northern US. The study also produced a new web-based visualization system to analyze the results of the study. The visualization system included interactive choropleth maps and charts to depict spatiotemporal changes in various climate thresholds over time.
- Top Cited Papers
- 30 June 2015
- Published by American Geophysical Union (AGU) in Reviews of Geophysics
- Vol. 53 (2) , 545-591
A review of tropical cyclone‐generated storm surges: Global data sources, observations, and impacts
Tropical cyclone‐generated storm surges are among the world's most deadly and destructive natural hazards. This paper provides the first comprehensive global review of tropical storm surge data sources, observations, and impacts while archiving data in SURGEDAT, a global database. Available literature has provided data for more than 700 surge events since 1880, the majority of which are found in the western North Atlantic (WNA), followed by Australia/Oceania, the western North Pacific (WNP), and the northern Indian Ocean (NIO). The Bay of Bengal (BOB) in the NIO consistently observes the world's highest surges, as this subbasin averages five surges ≥5 m per decade and has observed credible storm tide levels reaching 13.7 m. The WNP observes the highest rate of low‐magnitude surges, as the coast of China averages 54 surges ≥1 m per decade, and rates are likely higher in the Philippines. The U.S. Gulf Coast observes the second highest frequency of both high‐magnitude (≥5 m) and low‐magnitude (≥1 m) surges. The BOB observes the most catastrophic surge impacts, as 59% of global tropical cyclones that have killed at least 5000 people occurred in this basin. The six deadliest cyclones in this region have each killed at least 140,000 people, and two events have killed 300,000. Storm surge impacts transportation, agriculture, and energy sectors in the WNA. Oceania experiences long‐term impacts, including contamination of fresh water and loss of food supplies, although the highest surges in this region are lower than most other basins.
- 30 June 2015
- Published by American Geophysical Union (AGU) in Reviews of Geophysics
- Vol. 53 (2) , 545-591
Top Cited PapersA review of tropical cyclone‐generated storm surges: Global data sources, observations, and impacts
Tropical cyclone‐generated storm surges are among the world's most deadly and destructive natural hazards. This paper provides the first comprehensive global review of tropical storm surge data sources, observations, and impacts while archiving data in SURGEDAT, a global database. Available literature has provided data for more than 700 surge events since 1880, the majority of which are found in the western North Atlantic (WNA), followed by Australia/Oceania, the western North Pacific (WNP), and the northern Indian Ocean (NIO). The Bay of Bengal (BOB) in the NIO consistently observes the world's highest surges, as this subbasin averages five surges ≥5 m per decade and has observed credible storm tide levels reaching 13.7 m. The WNP observes the highest rate of low‐magnitude surges, as the coast of China averages 54 surges ≥1 m per decade, and rates are likely higher in the Philippines. The U.S. Gulf Coast observes the second highest frequency of both high‐magnitude (≥5 m) and low‐magnitude (≥1 m) surges. The BOB observes the most catastrophic surge impacts, as 59% of global tropical cyclones that have killed at least 5000 people occurred in this basin. The six deadliest cyclones in this region have each killed at least 140,000 people, and two events have killed 300,000. Storm surge impacts transportation, agriculture, and energy sectors in the WNA. Oceania experiences long‐term impacts, including contamination of fresh water and loss of food supplies, although the highest surges in this region are lower than most other basins.
- Open Access
- 11 June 2013
- Published by American Geophysical Union (AGU) in Eos
- Vol. 94 (24) , 213-214
A Global Database of Tropical Storm Surges
Tropical cyclone–generated storm surges are among the world's most deadly and costly natural disasters. The destructive nature of this hazard was clearly seen last fall, as Hurricane Sandy generated a devastating storm surge along the mid‐Atlantic coast. The storm killed 147 people and caused approximately $50 billion in economic losses [Blake et al., 2012].
- 11 June 2013
- Published by American Geophysical Union (AGU) in Eos
- Vol. 94 (24) , 213-214
Open AccessA Global Database of Tropical Storm Surges
Tropical cyclone–generated storm surges are among the world's most deadly and costly natural disasters. The destructive nature of this hazard was clearly seen last fall, as Hurricane Sandy generated a devastating storm surge along the mid‐Atlantic coast. The storm killed 147 people and caused approximately $50 billion in economic losses [Blake et al., 2012].
- 13 November 2012
- Published by American Society of Civil Engineers (ASCE)
- p. 715-740
Storm Surge Return Periods for the United States Gulf Coast
This study estimates tropical cyclone-generated storm surge levels for 2-year, 5-year, 10-year, 25-year, 50-year and 100-year return periods along the U.S. Gulf Coast, using SURGEDAT, a Gulf of Mexico storm surge database, for statistical analysis. Return periods are calculated using the Pareto, Gumbel and Beta-P distributions, and the Huff-Angel and Southern Regional Climate Center (SRCC) linear regression methods. The SRCC method, which fit best with the data, estimated basin-wide surge levels of 8.20 meters for the 100-year return period. The Southeast Louisiana/ Mississippi Zone generated the highest surge levels out of 10 sub-regions in the basin. Surge levels in this zone ranged were 7.67 meters for the 100-year return period. The lowest surge levels in the region were estimated along the West Coast of Florida. These results are useful for emergency management and disaster science personnel, planners, decision makers and the scientific research community.
- 13 November 2012
- Published by American Society of Civil Engineers (ASCE)
- p. 715-740
Storm Surge Return Periods for the United States Gulf Coast
This study estimates tropical cyclone-generated storm surge levels for 2-year, 5-year, 10-year, 25-year, 50-year and 100-year return periods along the U.S. Gulf Coast, using SURGEDAT, a Gulf of Mexico storm surge database, for statistical analysis. Return periods are calculated using the Pareto, Gumbel and Beta-P distributions, and the Huff-Angel and Southern Regional Climate Center (SRCC) linear regression methods. The SRCC method, which fit best with the data, estimated basin-wide surge levels of 8.20 meters for the 100-year return period. The Southeast Louisiana/ Mississippi Zone generated the highest surge levels out of 10 sub-regions in the basin. Surge levels in this zone ranged were 7.67 meters for the 100-year return period. The lowest surge levels in the region were estimated along the West Coast of Florida. These results are useful for emergency management and disaster science personnel, planners, decision makers and the scientific research community.
- 29 May 2012
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Knowledge and Data Engineering
- Vol. 25 (6) , 1354-1367
On Identifying Critical Nuggets of Information during Classification Tasks
In large databases, there may exist critical nuggets-small collections of records or instances that contain domain-specific important information. This information can be used for future decision making such as labeling of critical, unlabeled data records and improving classification results by reducing false positive and false negative errors. This work introduces the idea of critical nuggets, proposes an innovative domain-independent method to measure criticality, suggests a heuristic to reduce the search space for finding critical nuggets, and isolates and validates critical nuggets from some real-world data sets. It seems that only a few subsets may qualify to be critical nuggets, underlying the importance of finding them. The proposed methodology can detect them. This work also identifies certain properties of critical nuggets and provides experimental validation of the properties. Experimental results also helped validate that critical nuggets can assist in improving classification accuracies in real-world data sets.
- 29 May 2012
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Knowledge and Data Engineering
- Vol. 25 (6) , 1354-1367
On Identifying Critical Nuggets of Information during Classification Tasks
In large databases, there may exist critical nuggets-small collections of records or instances that contain domain-specific important information. This information can be used for future decision making such as labeling of critical, unlabeled data records and improving classification results by reducing false positive and false negative errors. This work introduces the idea of critical nuggets, proposes an innovative domain-independent method to measure criticality, suggests a heuristic to reduce the search space for finding critical nuggets, and isolates and validates critical nuggets from some real-world data sets. It seems that only a few subsets may qualify to be critical nuggets, underlying the importance of finding them. The proposed methodology can detect them. This work also identifies certain properties of critical nuggets and provides experimental validation of the properties. Experimental results also helped validate that critical nuggets can assist in improving classification accuracies in real-world data sets.