The U.S. Geological Survey (USGS), in cooperation with the New Hampshire Department of Environmental Services (NHDES) and the Department of Health and Human Services (NHDHHS), has developed data to assess the effects of short- and long-term climate change on hydrology in New Hampshire. A USGS Scientific Investigations Report (SIR) documents the datasets developed by the USGS. The data presented in this data release represent future hydrologic climate projections developed using a calibrated USGS Precipitation Runoff Modeling System (PRMS) model using precipitation and air temperature inputs from five general circulation models (GCMs) for two future climate scenarios for the period 2009 to 2099. The data sets include simulated current and future streamflow, groundwater recharge, and snowfall output datasets. Average monthly streamflow time series data sets are provided for 21 streamgages in New Hampshire, 14 of which also provide daily streamflow time series, Average monthly groundwater recharge and snowfall time series for the same reference time frame and future time frame are also provided for each of the 467 hydrologic response units (HRUs) that compose the model.

The U.S. Geological Survey has developed tools for projecting twenty-first century climate and hydrologic risk in Massachusetts in collaboration with Cornell University and Tufts University. These tools included a Stochastic Weather Generator (SWG). Output from the SWG is in this data release. The release includes daily precipitation and minimum and maximum air temperature for a 64-year period in the Nashua River watershed (that includes the Squannacook River) in Massachusetts and New Hampshire. There are 100 ensembles from the SWG for warming scenarios of 0 to 8 degrees Celsius in 0.5-degree increments. The SWG data were converted to a format utilized by the Precipitation-Runoff Modeling System (PRMS; https://www.usgs.gov/software/precipitation-runoff-modeling-system-prms) and input to a PRMS model for the Squannacook River watershed. The PRMS input and output files for the 100 ensembles of each of the 17 warming scenarios are also included in this data release. The 1,700 PRMS output files were utilized by a Stochastic Watershed Modeling tool to correct modeling biases that are inherent with a deterministic model such as PRMS. This data release includes the output from this Stochastic Watershed Model (SWM). For each of the 100 ensembles, the SWM was used to generate 10,000 ensembles, resulting in 1 million ensembles of 64-year periods for each of the warming scenarios. For each ensemble, streamflow characteristics of the annual maximum daily discharge at the 2-, 5-, 10-, 25-, 50-, 100-, and 500-year recurrence interval and of the annual 7-day low flow at the 2- and 10-year recurrence interval were determined.

The U.S. Geological Survey has developed tools for projecting twenty-first century climate and hydrologic risk in Massachusetts in collaboration with Cornell University and Tufts University. These tools included a Stochastic Weather Generator (SWG). Output from the SWG is in this data release. The release includes daily precipitation and minimum and maximum air temperature for a 64-year period in the Nashua River watershed (that includes the Squannacook River) in Massachusetts and New Hampshire. There are 100 ensembles from the SWG for warming scenarios of 0 to 8 degrees Celsius in 0.5-degree increments. The SWG data were converted to a format utilized by the Precipitation-Runoff Modeling System (PRMS; https://www.usgs.gov/software/precipitation-runoff-modeling-system-prms) and input to a PRMS model for the Squannacook River watershed. The PRMS input and output files for the 100 ensembles of each of the 17 warming scenarios are also included in this data release. The 1,700 PRMS output files were utilized by a Stochastic Watershed Modeling tool to correct modeling biases that are inherent with a deterministic model such as PRMS. This data release includes the output from this Stochastic Watershed Model (SWM). For each of the 100 ensembles, the SWM was used to generate 10,000 ensembles, resulting in 1 million ensembles of 64-year periods for each of the warming scenarios. For each ensemble, streamflow characteristics of the annual maximum daily discharge at the 2-, 5-, 10-, 25-, 50-, 100-, and 500-year recurrence interval and of the annual 7-day low flow at the 2- and 10-year recurrence interval were determined.

Climate change information simulated by global climate models is downscaled using statistical methods to translate spatially course regional projections to finer resolutions needed by researchers and managers to assess local climate impacts. Several statistical downscaling methods have been developed over the past fifteen years, resulting in multiple datasets derived by different methods. We apply a simple monthly water-balance model (MWBM) to demonstrate how the differences among these datasets result in disparate projections of snow loss and future changes in runoff. We apply the MWBM to six statistically downscaled datasets for 14 general circulation models (GCMs) from the Climate Model Intercomparison Program Phase 5 (CMIP5) for the RCP 8.5 emission scenario (1950 - 2099). The statistically downscaled datasets are as follows: BCCA: Bias Corrected Constructed Analogs (Reclamation, 2013) BCSD-C: Bias Corrected Spatial Disaggregation (Reclamation, 2013) BCSD-F: Bias Corrected Spatial Disaggregation (Thrasher et al., 2013) LOCA: Localized Constructed Analogs (Pierce et al., 2014) MACA-L: Multivariate Adaptive Constructed Analogs (Abatzoglou & Brown, 2012, bias corrected by Livneh et al., 2013) MACA-M: Multivariate Adaptive Constructed Analogs (Abatzoglou & Brown, 2012, bias corrected by METDATA, Abatzoglou, 2013) Users interested in the downscaled temperature and precipitation files are referred to the dataset home pages: BCCA, BCSD-C: http://gdo-dcp.ucllnl.org/downscaled_cmip_projections/dcpInterface.html BCSD-F: https://cds.nccs.nasa.gov/nex/ LOCA: http://loca.ucsd.edu/ MACA-L, MACA-M: http://maca.northwestknowledge.net The GCMs are the following: bcc-csm1-1, CanESM2, CNRM-CM5, CSIRO-Mk3-6-0, GFDL-ESM2G, GFDL-ESM2M, inmcm4, IPSL-CM5A-LR, IPSL-CM5A-MR, MIROC-ESM, MIROC-ESM-CHEM, MIROC5, MRI-CGCM3, NorESM1-M

This data release supports the study by Sexstone and others (2019) and contains simulation output from a hydrological modeling experiment using a specific calibration of the conterminous United States (CONUS) application of the Precipitation-Runoff Modeling System (PRMS) (Hay, 2019) as implemented in the National Hydrologic Model (NHM) infrastructure (Regan and others, 2018). The by hydrologic response unit (byHRU) calibrated, baseline version of the NHM-PRMS (Hay, 2019) was used to evaluate the sensitivity of simulated runoff to the representation of snow depletion curves (SDCs) within the NHM-PRMS across the CONUS. The model experiment consisted of seven NHM-PRMS model simulations using the calibrated NHM-PRMS model parameters from Hay (2019). For each of the model simulations, the calibrated SDCs (Hay, 2019) were replaced with a single derived SDC derived based on a lognormal probability distribution function and assigned snow water equivalent coefficient of variation (CV) value. The seven CV values ranged from 0.1 to 2.0. Each of the simulations were completed at a daily time-step over a 14-year period (water years 2003 – 2016). Detailed methods and results are provided in Sexstone and others (2019). The SDC parameters used in this model experiment are provided by this data release. Furthermore, the attached NHM-PRMS variable table lists the selected output variables included in this data release. The individual *.csv files follow a naming convention of nhru_variable name_CVX.X.csv. The variable names included are defined further in NHM-PRMS variable table. The “CVX.X” denotes the CV value that was used to derive the SDC for the model simulation. The structure of each output file includes a header line which labels the columns by the HRU identification number with each row providing daily outputs. An inventory of the files provided within this data release can be found below. This research used resources provided by the Core Science Analytics, Synthesis, & Libraries (CSASL) Advanced Research Computing (ARC) group at the U.S. Geological Survey. Inventory of data release: NHM-PRMS_SDC_study.xml (1 .xml file): FGDC-compliant metadata file for the data release files. SDC_params.csv (1 .csv file): Table of the seven snow depletion curve parameterizations used in the modeling study. NHM-PRMS_variable_table.docx (1 .docx file): Table describing the selected NHM-PRMS output variables provided in this data release. nhru_gwres_flow_CVX.X.csv (7 .csv files): NHM-PRMS groundwater discharge output for water years 2002 through 2016. Each of the 7 .csv files are labeled with the coefficient of variation (CV) value for the snow depletion curve that was used in the model simulation. nhru_hru_actet_CVX.X.csv (7 .csv files): NHM-PRMS actual evapotranspiration output for water years 2002 through 2016. Each of the 7 .csv files are labeled with the coefficient of variation (CV) value for the snow depletion curve that was used in the model simulation. nhru_hru_outflow_CVX.X.csv (7 .csv files): NHM-PRMS total flow output for water years 2002 through 2016. Each of the 7 .csv files are labeled with the coefficient of variation (CV) value for the snow depletion curve that was used in the model simulation. nhru_hru_ppt_CVX.X.csv (7 .csv files): NHM-PRMS precipitation output for water years 2002 through 2016. Each of the 7 .csv files are labeled with the coefficient of variation (CV) value for the snow depletion curve that was used in the model simulation. nhru_pkwater_equiv_CVX.X.csv (7 .csv files): NHM-PRMS snow water equivalent output for water years 2002 through 2016. Each of the 7 .csv files are labeled with the coefficient of variation (CV) value for the snow depletion curve that was used in the model simulation. nhru_potet_CVX.X.csv (7 .csv files): NHM-PRMS potential evapotranspiration output for water years 2002 through 2016. Each of the 7 .csv files are labeled with the coefficient of variation (CV) value for the snow depletion curve that was

This data release supports the study by Sexstone and others (2019) and contains simulation output from a hydrological modeling experiment using a specific calibration of the conterminous United States (CONUS) application of the Precipitation-Runoff Modeling System (PRMS) (Hay, 2019) as implemented in the National Hydrologic Model (NHM) infrastructure (Regan and others, 2018). The by hydrologic response unit (byHRU) calibrated, baseline version of the NHM-PRMS (Hay, 2019) was used to evaluate the sensitivity of simulated runoff to the representation of snow depletion curves (SDCs) within the NHM-PRMS across the CONUS. The model experiment consisted of seven NHM-PRMS model simulations using the calibrated NHM-PRMS model parameters from Hay (2019). For each of the model simulations, the calibrated SDCs (Hay, 2019) were replaced with a single derived SDC derived based on a lognormal probability distribution function and assigned snow water equivalent coefficient of variation (CV) value. The seven CV values ranged from 0.1 to 2.0. Each of the simulations were completed at a daily time-step over a 14-year period (water years 2003 – 2016). Detailed methods and results are provided in Sexstone and others (2019). The SDC parameters used in this model experiment are provided by this data release. Furthermore, the attached NHM-PRMS variable table lists the selected output variables included in this data release. The individual *.csv files follow a naming convention of nhru_variable name_CVX.X.csv. The variable names included are defined further in NHM-PRMS variable table. The “CVX.X” denotes the CV value that was used to derive the SDC for the model simulation. The structure of each output file includes a header line which labels the columns by the HRU identification number with each row providing daily outputs. An inventory of the files provided within this data release can be found below. This research used resources provided by the Core Science Analytics, Synthesis, & Libraries (CSASL) Advanced Research Computing (ARC) group at the U.S. Geological Survey. Inventory of data release: NHM-PRMS_SDC_study.xml (1 .xml file): FGDC-compliant metadata file for the data release files. SDC_params.csv (1 .csv file): Table of the seven snow depletion curve parameterizations used in the modeling study. NHM-PRMS_variable_table.docx (1 .docx file): Table describing the selected NHM-PRMS output variables provided in this data release. nhru_gwres_flow_CVX.X.csv (7 .csv files): NHM-PRMS groundwater discharge output for water years 2002 through 2016. Each of the 7 .csv files are labeled with the coefficient of variation (CV) value for the snow depletion curve that was used in the model simulation. nhru_hru_actet_CVX.X.csv (7 .csv files): NHM-PRMS actual evapotranspiration output for water years 2002 through 2016. Each of the 7 .csv files are labeled with the coefficient of variation (CV) value for the snow depletion curve that was used in the model simulation. nhru_hru_outflow_CVX.X.csv (7 .csv files): NHM-PRMS total flow output for water years 2002 through 2016. Each of the 7 .csv files are labeled with the coefficient of variation (CV) value for the snow depletion curve that was used in the model simulation. nhru_hru_ppt_CVX.X.csv (7 .csv files): NHM-PRMS precipitation output for water years 2002 through 2016. Each of the 7 .csv files are labeled with the coefficient of variation (CV) value for the snow depletion curve that was used in the model simulation. nhru_pkwater_equiv_CVX.X.csv (7 .csv files): NHM-PRMS snow water equivalent output for water years 2002 through 2016. Each of the 7 .csv files are labeled with the coefficient of variation (CV) value for the snow depletion curve that was used in the model simulation. nhru_potet_CVX.X.csv (7 .csv files): NHM-PRMS potential evapotranspiration output for water years 2002 through 2016. Each of the 7 .csv files are labeled with the coefficient of variation (CV) value for the snow depletion curve that was

This data release provides a set of Hydrological Simulation Program--Fortran (HSPF) model files representing 5 EPA-selected future climate change scenarios for each of two river basins: Taunton and Sudbury, in Massachusetts. Output from these models are intended for use as input to EPA Watershed Management Optimization Support Tool (WMOST) modeling. Climate scenarios, based on 2036-2065 change from 1975-2004 Representative Concentration Pathways (RCP) 4.5 and 8.5, model effects of air temperature and precipitation changes (in degrees F for air temperature, in percent for precipitation) made to the input historical meteorological time series 1975-2004. Taunton meteorological data is from T.F. Green Airport and the Sudbury meteorological data is from Worcester Regional Airport. Each set of climate scenario model files are derived from the original calibrated model files developed to support WMOST modeling (refer to Source Input fields in this metadata file).

This data release provides a set of Hydrological Simulation Program--Fortran (HSPF) model files representing 5 EPA-selected future climate change scenarios for the Upper Charles River Basin in Massachusetts. Output from these models are intended for use as input to EPA Watershed Management Optimization Support Tool (WMOST) modeling. Climate scenarios, based on 2036-2065 changes from 1975-2004 for Representative Concentration Pathways (RCP) 4.5 and 8.5, model the effects of air temperature and precipitation changes (in degrees F for air temperature, in percent for precipitation) made to the input historical meteorological time series for 1975-2004. Meteorological data are from Boston Airport (Boston, MA), T.F. Green Airport (Providence, RI), and Worcester Regional Airport (Worcester, MA). Each set of climate scenario model files are derived from the original calibrated model files created by the Charles River Watershed Association to develop Total Maximum Daily Loads (TMDLs) for nutrients, and modified by USGS to support WMOST modeling (refer to Source Input fields in this metadata file).

This data release provides a set of Hydrological Simulation Program--Fortran (HSPF) model files representing 5 EPA-selected future climate change scenarios for the Upper Charles River Basin in Massachusetts. Output from these models are intended for use as input to EPA Watershed Management Optimization Support Tool (WMOST) modeling. Climate scenarios, based on 2036-2065 changes from 1975-2004 for Representative Concentration Pathways (RCP) 4.5 and 8.5, model the effects of air temperature and precipitation changes (in degrees F for air temperature, in percent for precipitation) made to the input historical meteorological time series for 1975-2004. Meteorological data are from Boston Airport (Boston, MA), T.F. Green Airport (Providence, RI), and Worcester Regional Airport (Worcester, MA). Each set of climate scenario model files are derived from the original calibrated model files created by the Charles River Watershed Association to develop Total Maximum Daily Loads (TMDLs) for nutrients, and modified by USGS to support WMOST modeling (refer to Source Input fields in this metadata file).

The U.S. Geological Survey in cooperation with the Federal Emergency Management Agency has conducted a study to evaluate potential changes to1-percent annual exceedance probability streamflows. The study was conducted using the Precipitation Runoff Modeling System (PRMS). Climate inputs to the model of temperature and precipitation were scaled to anticipated changes that could occur in 2030, 2050, and 2100 based on global climate models. The output from the models were used to characterize the 1-percent AEP streamflows for the years 2030, 2050, and 2100 and compare the results to baseline conditions, 1950-2015. The data include the model input and output and spatial data for model referencing. Scripts for processing PRMS output to obtain final results are also provided.