Current estimates of the magnitude and frequency of floods at gaged and ungaged stream sites are critical for assessing flood risk, delineating flood zones, designing hydraulic structures, and managing flood plains. The Connecticut Department of Transportation collaborated with U.S. Geological Survey (USGS) in a study to improve the flood-frequency estimates in Connecticut and develop regional regression equations for estimating annual exceedance probability discharges at ungaged sites in Connecticut. The results of the study are found in Scientific Investigations Report (add link). This companion data release consists of data compiled and used for the flood-frequency analysis of annual peak flows and development of regional regression equations for Connecticut. The data release is composed of four tables that include: 1) descriptions and locations of streamgages for flood frequency analysis of annual peak flows; 2) updated flood discharges for 152 streamgages for the 50-, 20-, 10-, 4-, 2-, 1-, 0.5-, and 0.2-percent annual exceedance probabilities (AEP) using annual peak data through water year 2015; 3) representation of peak flow data using perception thresholds and interval data in the flood frequency analysis to describe uncertainty and represent missing and historical records; and 4) basin and climatic characteristics that were included as explanatory variables in the regression equations.

The U.S. Geological Survey (USGS), in cooperation with Connecticut Department of Transportation, completed a study to improve flood-frequency estimates in Connecticut. This companion data release is a Microsoft Excel workbook for: (1) computing flood discharges for the 50- to 0.2-percent annual exceedance probabilities from peak-flow regression equations, and (2) computing additional prediction intervals, not available through the USGS StreamStats web application. The current StreamStats application (version 4) only computes the 90-percent prediction interval for stream sites in Connecticut. The Excel workbook can be used to compute the 70-, 80-, 90-, 95-, and 99-percent prediction intervals. The prediction interval provides upper and lower limits of the estimated flood discharge with a certain probability, or level of confidence in the accuracy of the estimate. The standard error of prediction for the Connecticut peak-flow regression equations ranged from 26.3 to 45.0 percent (Ahearn and Hodgkins, 2020). The Excel workbook consists of four worksheets. The worksheets provide an overview of how the application works; input and output tables of the explanatory variables and flood discharges, and graphical display of the results; and the computational formulas used to estimate the flood discharges and prediction intervals.

The U.S. Geological Survey (USGS), in cooperation with Connecticut Department of Transportation, completed a study to improve flood-frequency estimates in Connecticut. This companion data release is a Microsoft Excel workbook for: (1) computing flood discharges for the 50- to 0.2-percent annual exceedance probabilities from peak-flow regression equations, and (2) computing additional prediction intervals, not available through the USGS StreamStats web application. The current StreamStats application (version 4) only computes the 90-percent prediction interval for stream sites in Connecticut. The Excel workbook can be used to compute the 70-, 80-, 90-, 95-, and 99-percent prediction intervals. The prediction interval provides upper and lower limits of the estimated flood discharge with a certain probability, or level of confidence in the accuracy of the estimate. The standard error of prediction for the Connecticut peak-flow regression equations ranged from 26.3 to 45.0 percent (Ahearn and Hodgkins, 2020). The Excel workbook consists of four worksheets. The worksheets provide an overview of how the application works; input and output tables of the explanatory variables and flood discharges, and graphical display of the results; and the computational formulas used to estimate the flood discharges and prediction intervals.

These data include flood-frequency data for the 50-, 20-, 10-, 4-, 2-, 1-, 0.5-, and 0.2-percent annual exceedance probabilities (pfqprocessed.zip) and the basin characteristics (siteinfoVT.txt) for 156 streamgages in Vermont and adjacent areas in surrounding states. These data were used as input to the regression model developed for estimating the magnitude of floods at the selected annual exceedance probabilities on ungaged, unregulated, rural streams in Vermont. The model code is provided in R script language (R Core Team, 2024) and utilized the generalized-least-squares regression tools in the WREG package (Farmer, 2021) to develop the regression equations to estimated flood discharge in Vermont. The model output (VT_Regression_Equations.pdf) provides code and the regression equations. Farmer, W.H., 2021, WREG—Weighted least squares regression for streamflow frequency statistics (ver. 3.0): U.S. Geological Survey software release, accessed January 31, 2025, at https://doi.org/10.5066/P9ZCGLI1. R Core Team, 2024, R: A language and environment for statistical computing version 4.4.0: R Foundation for Statistical Computing, Vienna, Austria, accessed July 18, 2024 at, https://www.R-project.org.

These data include annual instantaneous maximum discharge record for 156 streams and rivers in Vermont and adjacent areas of New Hampshire, Massachusetts, and New York through the 2023 water year. The data also include flood-frequency estimates for the 50-, 20-, 10-, 4-, 2-, 1-, 0.5-, and 0.2-percent annual exceedance probabilities that were analyzed using the discharge data. These flood-frequency data were used to develop regression equations for estimating the magnitude of floods at the selected annual exceedance probabilities on ungaged, unregulated, rural streams in Vermont. The regression equations use the basin characteristics of drainage area, percentage of wetland area in the basin, and the basinwide mean of the annual precipitation as explanatory variables. These basin characteristics for the streamgages and the Geographic Information System datasets for computing the basin characteristics are also in this data release. This data release is structured with the streamgage data, the input and output files for the version 7.5.1 PeakFQ software (U.S. Geological Survey, 2024) used to conduct flood-frequency analysis, the flood-frequency results, the streamgage basin characteristics, and the geographic information system datasets used to determine the basin characteristics on the main page. There is one sub-page which is used as a model archive for the regression analysis. U.S. Geological Survey, 2024, Water resources application software, PeakFQ, accessed March 4, 2024, at http://water.usgs.gov/software/peakfq.html.

The U.S. Geological Survey (USGS), in cooperation with the New Mexico Department of Transportation, estimated the magnitude and frequency of floods corresponding to the 50-, 20-, 10-, 4-, 2-, 1-, 0.5-, and 0.2-percent annual exceedance probabilities (AEPs; otherwise known as the 2-, 5-, 10-, 25-, 50-, 100-, 200-, and 500-year floods, respectively) for 346 selected streamgages in New Mexico and parts of Arizona, Colorado, Oklahoma, Texas, and Utah using data through water year 2020. An updated regional flood skew, -0.145, standard error 0.454, was computed for the study area. Regression equations were developed which can be used to estimate the magnitude and frequency of floods at ungaged locations on unregulated streams in the study area. The methods and results of the study are published in the parent report (Bell and others, 2022, https://doi.org/10.5066/XXXXXXXX). For the 346 selected streamgages, this dataset includes peak-flow (*.pkf) and specification (*.psf), output (*.PRT), and export (*.EXP) files from version 7.3 of USGS PeakFQ software (Veilleux and others, 2014; Flynn and others, 2006). Within PeakFQ software, the Expected Moments Algorithm (EMA) was used to conduct frequency analyses to estimate stream discharges corresponding to the 0.5, 0.2, 0.1, 0.04, 0.02, 0.01, 0.005, and 0.002 AEPs. When appropriate, the updated regional skew was used to weight the at-site skew in the frequency analyses. Results of the frequency analyses were used in generalized least-squares (GLS) regression to generate equations that predict discharges corresponding to selected AEPs at ungaged locations on streams in the study area (Bell and others, 2022).

This dataset contains site information and results of flood-frequency analysis for 422 streamflow gaging stations (streamgages) operated by the U.S. Geological Survey (USGS) in parts of the Upper Mississippi and Souris-Red-Rainy basins and surrounding areas in the United States. Annual peak-flow data from the 1844 - 2013 water years were used in the study (U.S. Geological Survey, 2024). Following federal guidelines for flood-frequency analysis (Bulletin 17C; England and others, 2018) and methods outlined in recent flood-frequency reports for the region (Eash and others, 2013; Southard and Veilleux, 2014; Levin and Sanocki, 2023; Sanocki and Levin, 2023), the Expected Moments Algorithm (EMA) was used in version 7.5.1 of USGS PeakFQ software (Veilleux and others, 2014; Flynn and others, 2006; https://water.ugsgs.gov/software/PeakFQ/) to conduct the analyses. Results of the analyses, specifically the at-site skew and its mean squared error, are intended for use in Bayesian weighted least-squares/Bayesian generalized least-squares (B-WLS/B-GLS) regression (Veilleux and Wagner, 2019; Veilleux and Wagner, 2021) to model regional skew for the study area. Peak-flow (.pkf), specification (.psf), output (.PRT), and export (.EXP) files from PeakFQ and a .csv file containing site information and selected results of flood-frequency analyses are provided.

This U.S. Geological Survey (USGS) data release contains batch formatted annual peak streamflow data (PkFlows_AllSites.txt) through the 2020 water year for six selected USGS streamgages (01321000, 01342797, 01343060, 01346000, 01347000, and 01348000) that recorded the flood of October 31 – November 3, 2019, which severely affected the Mohawk Valley and southern Adirondack region in central New York State. This data release also contains batch formatted specification (PkFlows_AllSites.psf) and output (PEAKFLOWS_ALLSITES.PRT) files from log-Pearson type III (LPIII) flood-frequency analysis of the annual peak streamflow data in version 7.4 of the USGS PeakFQ software (Flynn and others, 2006), which implements the Bulletin 17C (England and others, 2018) methodology for determining flood flow frequency. A comma separated values file (FloodFrequencyEstimates.csv) with estimates of flood magnitudes for selected annual exceedance probabilities from three different types of flood-frequency analysis (LPIII, regional regression equations, and weighted LPIII) is also included. The basins of the selected streamgages include the Sacandaga River basin (located within the upper Hudson River basin), and the East and West Canada Creek basins (located within the Mohawk River basin). Annual peak streamflow data were obtained from the USGS National Water Information System (U.S. Geological Survey, 2016). Considerations that informed the PeakFQ specifications for all six streamgages (including the effect of regulation, differences in station and regional skew, and other considerations) are discussed in Graziano and others (2024). Regional regression equation estimates were obtained for the locations of the four streamgages where streamflow is unregulated (01321000, 01342797, 01343060, and 01348000) using the StreamStats web application (U.S. Geological Survey, 2019), which presently utilizes equations in Lumia and others (2006). Weighted LPIII estimates (for the same four streamgages where streamflow is unregulated) were computed by weighting the (at-site) LPIII estimates with the regional regression equation estimates (Lumia and others (2006), equation 3). The weighted LPIII estimates are considered the most accurate of the three types of flood-frequency estimates. For this analysis, stationarity in the annual peak streamflow records was assumed after visual inspection of the records indicated no clear trends.

This U.S. Geological Survey (USGS) data release contains batch formatted annual peak streamflow data (PkFlows_AllSites.txt) through the 2020 water year for six selected USGS streamgages (01321000, 01342797, 01343060, 01346000, 01347000, and 01348000) that recorded the flood of October 31 – November 3, 2019, which severely affected the Mohawk Valley and southern Adirondack region in central New York State. This data release also contains batch formatted specification (PkFlows_AllSites.psf) and output (PEAKFLOWS_ALLSITES.PRT) files from log-Pearson type III (LPIII) flood-frequency analysis of the annual peak streamflow data in version 7.4 of the USGS PeakFQ software (Flynn and others, 2006), which implements the Bulletin 17C (England and others, 2018) methodology for determining flood flow frequency. A comma separated values file (FloodFrequencyEstimates.csv) with estimates of flood magnitudes for selected annual exceedance probabilities from three different types of flood-frequency analysis (LPIII, regional regression equations, and weighted LPIII) is also included. The basins of the selected streamgages include the Sacandaga River basin (located within the upper Hudson River basin), and the East and West Canada Creek basins (located within the Mohawk River basin). Annual peak streamflow data were obtained from the USGS National Water Information System (U.S. Geological Survey, 2016). Considerations that informed the PeakFQ specifications for all six streamgages (including the effect of regulation, differences in station and regional skew, and other considerations) are discussed in Graziano and others (2024). Regional regression equation estimates were obtained for the locations of the four streamgages where streamflow is unregulated (01321000, 01342797, 01343060, and 01348000) using the StreamStats web application (U.S. Geological Survey, 2019), which presently utilizes equations in Lumia and others (2006). Weighted LPIII estimates (for the same four streamgages where streamflow is unregulated) were computed by weighting the (at-site) LPIII estimates with the regional regression equation estimates (Lumia and others (2006), equation 3). The weighted LPIII estimates are considered the most accurate of the three types of flood-frequency estimates. For this analysis, stationarity in the annual peak streamflow records was assumed after visual inspection of the records indicated no clear trends.

A major rain event caused catastrophic flooding from July 9 through 12, 2023 in various portions of the State of Vermont, resulting in millions of dollars of damage. This is one of 3 datasets in a data release that documents the flood data and analyses. This part of the data release contains the peak streamflow and lake level data resulting from the July 9 through 12, 2023 flooding as well as the flood-frequency analysis that was completed implementing the Bulletin 17C guidelines (https://doi.org/10.3133/tm4B5). Flood flows were estimated for the 50-, 20-, 10-, 4-, 2-, 1-, 0.5-, and 0.2-percent annual exceedance probabilities (AEPs) for all streamgages operated in Vermont by the USGS where data was collected. This data release presents frequency analyses for 80 streamgages and 2 lake gages in Vermont and an additional 6 sites where peak streamflows were determined using indirect measurement techniques. The data release contains the input and output files for the version 7.5.1 PeakFQ software (https://water.usgs.gov/software/PeakFQ/; Veilleux and others, 2014) used to conduct flood-frequency analyses using the expected moments algorithm (England and others, 2018). The peak-flow files used as input to PeakFQ were obtained from the USGS National Water Information System (NWIS) database (https://nwis.waterdata.usgs.gov/usa/nwis/peak) and incorporate annual peak flows through water year 2023. Results of the flood-frequency analyses at streamgages that had unregulated peak streamflows were weighted with regional flood-frequency regression equation results (Olson, 2014). The flood frequency for the two lake gages was estimated using HEC-SSP version 2.2 (https://www.hec.usace.army.mil/software/hec-ssp/) because the software allows flood frequency analysis without logarithmic transformation of the data. In addition to the input and output for the PeakFQ software, this data release has four Excel tables summarizing the flood data, the flood frequency results, and a summary table of information describing the data and analysis for each streamgage.