List of scripts and datasets in this HydroShare resource

Datasets:

Scripts:

Scripts are listed below in three sequential categories:
1. acquisition and/or compilation of individual data sources
2. aggregation of multiple data sources into single dataset for analysis
3. analyses used to produce results, tables, and figures

All scripts have a .r extension and need to be run using the R language and environment for statistical computing available from: http://r-project.org.

Reproducibility notes:

To reproduce only analysis of compiled datasets, run scripts only the ANALYSIS block below, which require two csv files included in this HydroShare resource (wb_filtered.csv and watersheds.csv).

To reproduce all work, beginning with data acquisition, scripts are intended to be run in the order in which they appear in this Readme file. Some manual downloads are required (as described in each script).

Scripts

ACQUISITION AND/OR COMPILATION OF INDIVIDUAL DATA SOURCES

compilation_CAMELS.r

• inputs:

  • CAMELS watershed attributes (long-term or time invariant) from several text files included in the CAMELS data download:
    • camels_hydro.txt
    • camels_clim.txt
    • camels_geol.txt
    • camels_name.txt
    • camels_soil.txt
    • camels_vege.txt
  • CAMELS daily streamflow data, from the downloaded CAMELS dataset, in the filepath /basin_timeseries_v1p2_metForcing_obsFlow/basin_dataset_public_v1p2/usgs_streamflow/XX/*_streamflow_qc.txt, where XX=USGS region (2-digit number) and * is an 8-digit gauge ID
  • CAMELS forcing data (Daymet), from the downloaded CAMELS dataset, in the filepath basin_timeseries_v1p2_metForcing_obsFlow/basin_dataset_public_v1p2/basin_mean_forcing/daymet/XX/*_lump_cida_forcing_leap.txt, where XX=USGS region (2-digit number) and * is an 8-digit gauge ID
    
    

• outputs:

  • camels_chars.csv - watershed attributes (long-term or time invariant) from the CAMELS dataset
  • wb_1980_2014.csv - annual and seasonal water budget components and forcing data for the CAMELS period of record (1980-2014); note that seasonal values were not analyzed for this paper

compilation_Daymet.r

• inputs:

  • HCDN_nhru_final_671.shp - shapefile included with CAMELS data download in the folder /basin_set_full_res
  • monthly Daymet gridded prcp, tmin, tmax, srad, swe, and dayl - respectively, precipitation, min temperature, max temperature, solar radiation, snow water equivalent (which was not used to do missing data in the CAMELS dataset), and day length (used to estimate PET); all downloaded via R package daymetr.
    
    

• outputs:

  • camels_prcp_2014_2019.csv - precipitation for 2014-2019 from Daymet prcp
  • camels_tave_2014_2019.csv - mean temperature for 2014-2019 based on Daymet tmin and tmax
  • camels_pet_2014_2019.csv - estimated PET for 2014-2019 using the Hamon method, based on Daymet inputs

compilation_USGS.r

• inputs:

  • daily streamflow for each watershed's gauge, obtained and summarized using R package DataRetrieval
  • wb_1980_2014.csv - annual water budget components and forcing data for the CAMELS period of record (1980-2014); created by compilation_CAMELS.r
    
    

• outputs:

  • flow_2015_2019.csv - annual streamflow for each watershed for 2014-2019, summarized from USGS daily data

compilation_FIA.r

• inputs:

  • data_FIA.csv - compiled and formatted data based on outputs of tree live volume, net growth, growth per acre, and mortality, all based on queries of the online EVALIDator tool (https://apps.fs.usda.gov/Evalidator/evalidator.jsp) from 10 April 2021
    
    

• outputs:

  • forest_data.csv - a csv file including tree growth and mortality output from FIA's EVALIDator online tool
    
    

COMPILATION OF MULTIPLE DATA SOURCES

compilation_CAMELS_Daymet_USGS.r

• inputs:

  • wb_1980_2014.csv - annual and seasonal water budget components and forcing data for the CAMELS period of record (1980-2014); note that seasonal values were not analyzed for this paper
  • camels_prcp_2014_2019.csv - precipitation for 2014-2019 from Daymet prcp
  • camels_tave_2014_2019.csv - mean temperature for 2014-2019 based on Daymet tmin and tmax
  • camels_pet_2014_2019.csv - estimated PET for 2014-2019 using the Hamon method, based on Daymet inputs
  • flow_2015_2019.csv - annual streamflow for each watershed for 2014-2019, summarized from USGS daily data
    
    

• outputs:

  • wb_1980_2019.csv - annual water budget components (Q, P) and drivers (PET, T) for 1980-2019

watershed_compilation.r

• inputs:

  • wb_1980_2019.csv - annual water budget components and climatic drivers
  • camels_chars.csv - watershed attributes (long-term or time invariant) from the CAMELS dataset
  • camels_ann.csv - an intermediate dataset used to summarize long-term mean incoming solar radiation
  • forest_data.csv - a csv file including tree growth and mortality output from FIA's EVALIDator online tool
    
    

• outputs:

  • watersheds.csv - one of two main data files used for analysis; this one contains watershed characteristics compiled from CAMELS, FIA, and the extended record of Daymet + USGS gage data
  • wb_filtered - two of two main data files used for analysis; this one is a time series of water budget components and forcing data, and it has been filtered to exclude watersheds where runoff ratio > 1 in any year.
    
    

ANALYSIS - scripts that produce results, figures, and tables in the associated the manuscript (can be run in any order; each is independent)

Map_watersheds.r

• inputs:

  • wb_filtered.csv - annual water budget components and climatic drivers, filtered to exclude watersheds with runoff ratio > 1
  • watersheds.csv - compiled watershed characteristics from FIA data and CAMELS dataset + extended record
  • HCDN_nhru_final_671.shp - shapefile included with CAMELS data download in the folder /basin_set_full_res/
    
    

• outputs:

  • FIGURE: map of CAMELS watersheds used in this study
  • TABLE: summarizing a few topographic and climatic characteristics of all watersheds

Analysis_M-K_trend_test.r

• inputs:

  • wb_filtered.csv - annual water budget components and climatic drivers, filtered to exclude watersheds with runoff ratio > 1
  • watersheds.csv - compiled watershed characteristics from FIA data and CAMELS dataset + extended record
    
    

• outputs:

  • FIGURE:
    • boxplot of Kendall's tau for Q, Q/P, P, PET, and T for 2000-2019
    • maps of watersheds, symbolized as significant increasing trend, significant decreasing trend, or no significant trend

analysis_M-K_quadrant_figure.r

• inputs:

  • wb_filtered.csv - annual water budget components and climatic drivers, filtered to exclude watersheds with runoff ratio > 1
  • watersheds.csv - compiled watershed characteristics from FIA data and CAMELS dataset + extended record
    
    

• outputs:

  • FIGURE:
    • scatter plot of trend in Q/P (Kendall's tau) vs. net tree growth
    • Budyko plot of evaporative index (ET/P) vs. aridity index (PET/P)
    • boxplot of aridity index (PET/P) for each quadrant of the scatter plot
    • boxplot of incoming solar radiation for each quadrant of the scatter plot
    • map showing locations of watersheds in each quadrant of the scatter plot

analysis_timetrend_linear.r

• inputs:

  • wb_filtered.csv - annual water budget components and climatic drivers, filtered to exclude watersheds with runoff ratio > 1
  • watersheds.csv - compiled watershed characteristics from FIA data and CAMELS dataset + extended record
    
    

• outputs:

  • FIGURE: map showing watersheds where mean annual Q for 2010-2019 deviated significantly from mean annual Q for 2000-2010, as well as magnitude of deviation and disturbance status (disturbed vs. undisturbed)
  • TABLE: tabular summary of the number of watersheds where mean annual Q for 2010-2019 deviated from mean annual Q for 2000-2010, by disturbance status (disturbed vs. undisturbed) and signifance (yes/no)

analysis_regressn_w-veg.r

• inputs:

  • wb_filtered.csv - annual water budget components and climatic drivers, filtered to exclude watersheds with runoff ratio > 1
  • watersheds.csv - compiled watershed characteristics from FIA data and CAMELS dataset + extended record
    
    

• outputs:

  • FIGURES:

    • forest plot of standardized coefficients (how many SDs the response will change per SD change in the predictor)
    • partial regression plots of the five terms in the regression model
    • boxplot of delta-Q/Q1 by aridity (PET/P) and mortality level, based on observations
    • fixed effects plot of mortality and aridity on delta-Q

  • TABLES:

    • multiple regression coefficients, statistics, and p-values
    • predicted percent change in Q due to multiple values of mortality and aridity, with and without 1 degree C of warming