Input Files and Code for: Machine learning can accurately assign geologic basin to produced water samples using major geochemical parameters

As more hydrocarbon production from hydraulic fracturing and other methods produce large volumes of water, innovative methods must be explored for treatment and reuse of these waters. However, understanding the general water chemistry of these fluids is essential to providing the best treatment options optimized for each producing area. Machine learning algorithms can often be applied to datasets to solve complex problems. In this study, we used the U.S. Geological Survey’s National Produced Waters Geochemical Database (USGS PWGD) in an exploratory exercise to determine if systematic variations exist between produced waters and geologic environment that could be used to accurately classify a water sample to a given...

Categories: Data; Tags: Alabama, Alaska, Alaska Region, Arizona, Arkansas, All tags... California, Colorado, Colorado Plateau and Basin and Range Region, Eastern Region, Energy Resources, Florida, Geochemical analysis, Geochemistry, Georgia, Gulf Coast Region, Idaho, Illinois, Indiana, Iowa, Kansas, Kentucky, Louisiana, Lousiana, Machine learning, Maryland, Michigan, Mid Continent Region, Mississippi, Missouri, Montana, Naïve Bayes, Nebraska, Nevada, New Mexico, New York, North Dakota, Ohio, Oklahoma, Ontario, Oregon, Pacific Coast Region, Pennsylvania, Produced waters, Random Forest, Rocky Mountains and Northern Great Plains Region, Saskatchewan, South Dakota, Tennessee, Texas, USGS Science Data Catalog (SDC), Utah, Virginia, Washington, West Texas and Eastern New Mexico Region, West Virginia, Wyoming, geochemistry, k-Nearest Neighbors, Fewer tags

Summary of streamflow statistics for USGS streamgages in the southeastern United States: 1950 - 2010

This dataset contains statistical descriptions of observed daily-mean streamflow for 956 sites in the southeast United States. For each site, statistical descriptions are provided according to decade for up to six decades, beginning in 1950 (1950-59 calendar years) and ending with 2000 (2000 - 2009 calendar years) with no more than 7 missing values per year in total (continuous or noncontinuous). There are 40 statistical descriptions including 28 flow-duration curve values, 8 L-moments, and 4 describing the minimum, maximum, median flow for days not equal to zero, and number of zero-flow days. Site information is provided by decade - the number of rows per site varies from 1 to 6 depending on the number of...

Categories: Data; Types: Data, Map Service, OGC WFS Layer, OGC WMS Layer, OGC WMS Service, Shapefile; Tags: Alabama (AL), Arkansas, Florida, Georgia, Gulf Coast, All tags... Gulf Coast Region, Gulf of Mexico, Louisiana, Mississippi, RESTORE, Southeastern United States, Tennessee, Texas, USGS Science Data Catalog (SDC), environmental flow, network optimization, reference hydrology, streamflow alteration, Fewer tags