An official website of the United States government
To increase geospatial awareness about local water resources, our team developed learning resources for the 150 km² Lake Sidney Lanier reservoir located in North Georgia, USA. The reservoir is vital for hydroelectric power generation, recreation, tourism, and consumptive uses. Using geospatial analysis in Google Earth Engine (GEE), we analyzed precipitation trends in the watershed using Climate Hazards Group InfraRed Precipitation with Station (CHIRPS) data. We also quantified expansion and contraction of reservoir surface area using Landsat-derived Global Surface Water data. As Lake Sidney Lanier is a managed reservoir, surface water extent fluctuations are related to climatic variables, consumptive use, and hydropower generation. Water temperature varies based on seasonality, water depth, water clarity, and lake stratification. Changing temperature dynamics affect ecosystem health and determine other important water quality parameters such as dissolved oxygen concentrations. Landsat 8 Thermal Infrared Sensor (TIRS) data were used to examine temperature trends over multiple years and investigate the timing of lake stratification and mixing. Highly turbid waters are associated with pollutants and lower water quality and can affect ecosystem productivity by minimizing sunlight penetration into the water column. Sentinel 2 MSI data were processed using a turbidity algorithm to analyze temporal trends and spatial correlations with reservoir inflows. Finally, high concentrations of chlorophyll a were used as a proxy to identify harmful algal blooms. The spatial differences in headwaters and near-dam locations were examined and near real-time satellite data were explored for potential development of early-warning systems to protect ecosystem and human health.
more »
« less
A Supervised Learning Approach to Water Quality Parameter Prediction and Fault Detection
Water quality parameters such as dissolved oxygen and turbidity play a key role in policy decisions regarding the maintenance and use of the nation's major bodies of water. In particular, the United States Geological Survey (USGS) maintains a massive suite of sensors throughout the nation's waterways that are used to inform such decisions, with all data made available to the public. However, the corresponding measurements are regularly corrupted due to sensor faults, fouling, and decalibration, and hence USGS scientists are forced to spend costly time and resources manually examining data to look for anomalies. We present a method of automatically detecting such events using supervised machine learning. We first present an extensive study of which water quality parameters can be reliably predicted, using support vector machines and gradient boosting algorithms for regression. We then show that the trained predictors can be used to automatically detect sensor decalibration, providing a system that could be easily deployed by the USGS to reduce the resources needed to maintain data fidelity.
more »
« less
- Award ID(s):
- 1758006
- PAR ID:
- 10393024
- Date Published:
- Journal Name:
- 2018 IEEE International Conference on Big Data (Big Data)
- Page Range / eLocation ID:
- 2511 to 2514
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
An ongoing component of the Baltimore urban long-term ecological research (LTER) project (Baltimore Ecosystem Study, BES) is the use of the watershed approach and monitoring of stream water quality to evaluate the integrated ecosystem functioning of Baltimore. The LTER research has focused on the Gwynns Falls watershed, which spans a gradient from highly urban, urban-residential, and suburban zones. In addition, a forested watershed serves as a reference. The long-term sampling network includes four longitudinal sampling sites along the Gwynns Falls mainstem, as well as several small (40-100 ha) watershed within or near the Gwynns Falls, providing data on water quality in different land use zones of the watersheds. Each study site is continuously monitored for discharge and is sampled weekly for water chemistry. Those data are available elsewhere on the BES website. We are interested in studying the bioreactivity of streams in our watersheds in an attempt to quantify how streams themselves may affect or be affected by water quality. To assess the bioreactivity of streams, we measure whole stream metabolism, which is an integrative metric which quantifies the production and consumption of energy by a stream ecosystem. Stream metabolism represents how energy is created (primary production) and used (respiration) within a stream; it can be thought of as a stream breathing, with primary production being similar to an inhale, and respiration as an exhale. We are monitoring stream metabolism in each of our long-term water quality monitoring stations by deploying sensors that record dissolve oxygen and temperature of the stream every five minutes, and we also have deployed light sensors to record irradiance every five minutes at long-term BES water chemistry streams, which is needed for metabolism modeling. In addition, each dissolved oxygen sensor is located near a USGS gage which estimates discharge every 15 minutes. We used USGS manual discharge estimations linked with channel geometry measurements to develop a unique discharge-stream depth relationship (contact AJ Reisinger for details). The combination of the USGS discharge data and our discharge-depth relationship allows us to estimate average daily discharge and depth. We have included these data as well as dissolved oxygen, temperature, and PAR, allowing metabolism to be scaled on an areal basis. Primary production and respiration of streams integrate all biological activity in a stream, and therefore are good metrics to assess the state of an ecosystem. These metrics can also be used to predict other ecosystem functions. This dataset includes all information needed for whole-stream metabolism modeling using the streammetabolizer R package. Data will updated as it becomes available from the core stream study sites (see http://md.water.usgs.gov/BES for a detailed description of these sites).more » « less
-
ABSTRACT The United States SECURE Water Act states that a study of water use is critical for assessing impacts on water and ecological resources and forecasting whether or not available surface and groundwater supplies will meet future needs. The United States Geological Survey (USGS) plays a key role in the SECURE Water Act by providing nationally consistent information on water quantity, quality, and use. Water‐use data maintained by States and Territories are critical for the USGS water‐use estimation and modeling techniques that underlie these efforts. However, water‐use data availability has not been systematically assessed. This study addresses this gap through a survey of USGS Water Science Centers (WSCs). The results indicate that water‐use information varies in its content and level of detail across the United States. Spatially discrete and comprehensive information about water use, such as site‐specific consumptive use, withdrawals, diversions, return flows, and interbasin transfers, is not widely available to and/or shared between State and Territory water‐resource agencies and USGS WSCs. This article presents the survey results and discusses reported barriers to water‐use data availability and sharing, as well as potential implications of limited water‐use information. This study advances understanding of water‐use data availability and sharing and contributes to broader research on US water data governance.more » « less
-
The task of automatically assessing and adjusting image quality, when capturing face images using any band-specific camera sensor, can be achieved by eliminating a variety of acquisition parameters such as illumination. One such parameter related to image quality is sharpness. If it is not accurately estimated during data collection, it may affect the quality of the overall face image dataset and thus face recognition accuracy. While manually focusing each camera on the target (human face) can result in sharp looking face images, the process can be cumbersome for the operators and the subjects and, thus, it increases data collection acquisition time. In this work, we developed an electromechanical based system that automatically assesses face image sharpness, prior to capture rather than necessitating post-processing schemes. Various blur quality factors and constraints have been empirically evaluated, before determining the algorithmic steps of our proposed system. This paper discusses the implementation of this system in a live operating system.more » « less
-
Human activities are causing global change around the world including habitat destruction, invasive species in non-native ecosystems, overexploitation, pollution, and global climate change. While traditional monitoring has long been used to quantify and aid mitigation of global change,in-situautonomous sensors are being increasingly used for environmental monitoring. Sensors and sensor platforms that can be deployed in developed and remote areas and allow high-frequency data collection, which is critical for parameters that exhibit important short-term dynamics on the scale of days, hours, or minutes. In this article, we discuss the benefits ofin-situautonomous sensors in aquatic ecosystems as well as the many challenges that we have experienced over many years of working with these technologies. These challenges include decisions on sensor locations, sensor types, analytical specification, sensor calibration, sensor drift, the role of environmental conditions, sensor fouling, service intervals, cost of ownership, and data QA/QC. These challenges result in important tradeoffs when making decisions regarding which sensors to deploy, particularly when a network of sensors is desired to cover a large area. We also review recent advances in designing and building chemical-sensor platforms that are allowing researchers to develop the next-generation of autonomous sensors and the power of integrating multiple sensors into a network that provides increased insight into the dynamics of water quality over space and time. In the coming years, there will be an exponential growth in data related to aquatic sensing, which will be an essential part of global efforts to monitor and mitigate global change and its adverse impacts on society.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- https://doi.org/10.1109/BigData.2018.8622628
