README

Tyler Hampton 2020-11-11

Intro

This R Markdown README file accompanies Hampton et al. (2020): A novel Budyko-based approach to quantify post-forest-fire streamflow response and recovery timescales

Corresponding Author Information: Tyler B. Hampton Email: tyler.hampton@uwaterloo.ca; ORCID: https://orcid.org/0000-0003-1087-2059

This document details the scripts and data accompanying this Rproject

Scripts

r list.files(pattern=".R")

## [1] "Budyko1_Rproj.Rproj" "code_0.1.R"          "code_0.2.R"         
## [4] "code_1.0.R"          "README.Rmd"

Note, these scripts rely on custom R packages developed by me, and posted on my github page (tylerbhampton).

r devtools::install_github("tylerbhampton/USGSreadR") devtools::install_github("tylerbhampton/budykoR")

“code_0.1.R”

This code outlines the first step of data retrieval used in this analysis. It outlines the use of climate data from PRISM (for citation see Supporting Information (SI) Section 1.2) including calculated gridded values for potential evapotranspiration (PET, See SI Section 1.2) to calculate average values per water-year for each catchment. This code is called from “code_0.2.R”

“code_0.2.R”

This code outlines the second step of data retrieval in this analysis. All inputs to this script are local to this R project except for shapefiles of USGS gages (from the GAGES dataset, see Table SI-3). This script applies the filters for catchment selection described in SI Section 1.1.

“code_1.0.R”

This is the primary script accompanying this publication. Running it in its entirety will process all the raw data, create figures in the figures folder and write all datasets to the outdata folder.

Note, the section in “code_1.0.R” for Supporting Information Section 2.1 (Figures SI 2) is turned off, as it does require external datasets from the CAMELS dataset which are too large to include with this Rproj. See filepaths in that code section for replication.

Input Files

r list.files(path="inputs")

## [1] "CAMELS"          "GAGES"           "MTBS"            "PRISM"          
## [5] "RiceEmanuel2019" "USGS_Qdata"

“USGS_Qdata”

This folder contains raw input files for streamflow data for gages in Southern California in the “WholeQseries” folder, and water-year summed values joined with climate data in the “DischargeAnnualSummaries” folder. These files are retrieved and generated by the dataRetrieval package which is called by my custom package USGSread. All files were retrieved in “code_0.2.R”. See “PRISM” folder for climate data and source. Files:

• “NWISdataX_00060.csv”: Daily Q Records

• “DischargeSummary_X.csv”: Annual Q Summary File

“GAGES”

This folder contains input files from the GAGES dataset (Falcone, 2011), which include watershed characteristics (climate, topology, land cover, etc.) for 9067 USGS stream gage watersheds in the conterminous United States. Files:

• “conterm_basinid.txt”

• “conterm_climate.txt”

• “conterm_lc06_basin.txt”

• “conterm_topo.txt”

This folder also contains a merged shapefile containing a subset of National Hydrography Dataset watersheds for Southern California used in this analysis (see origin in “code_0.2.R”). Files

• “southernCA_gages_data”

“MTBS”

This folder contains shapefiles and raster files for wildfires in Southern California from the MTBS dataset (Eidenshink et al., 2007). Files:

• “CA_firedata_all_bndy”: Merged shapefile of California wildfires (1984-2019).

• “X_dnbr6.dif”: Differenced Normalized Burn Ratio raster files for individual wildfires, showing burn severity.

“PRISM”

This folder contains csv files (one for each USGS gage) containing water-year-summed precipitation and potential evapotranspiration for WY 1982-2019. These files are generated according to “code_0.1.R” and called from “code_0.2.R” Files:

• “PRISMppet_bySite_X.csv”

RiceEmanuel2019

This folder contains supporting data from Rice & Emanuel (2019), who analyzed interannual variability in storage carryover in the water budgets of watersheds in the U.S. This data is called for Figure SI-1 in “code_1.0.R”

“CAMELS”

This folder contains inputs from the CAMELS dataset (Addor et al., 2017; See SI Section 2.1). We used gridded climate estimates from CAMELS discritized to watersheds to estimate how different potential evapotranspiration (PET) equations affected total annual PET values.

Output Files

r list.files(path="outdata")

##  [1] "Data_sCABudyko.csv"            "Figure2e_allburndata.csv"     
##  [3] "Figure2e_data.csv"             "Figure3_BudykoFitdata.csv"    
##  [5] "Figure4_MeansPValues.csv"      "Figure5_FlowDecomposition.csv"
##  [7] "Figure6_NLSbudykodata.csv"     "Figure6_Recoverydata.csv"     
##  [9] "Figure6_TimeSeriesData.csv"    "FigureSI2_camelPETcompare.csv"
## [11] "TableSI-1.csv"                 "TableSI-2.csv"

“Data_sCABudyko.csv” This file is generated in “code_0.2.R” and is a product of joining together GAGES catchment characteristics, all-time averages of flow and climate data, and estimates of Fu-type Budyko parameters for pre-fire (\<2004WY) water-year timeseries (using the budyko package).

“TableSI-1.csv” Data generated for Table SI-1 in the supporting information Generated in “PAPER TABLES” code section of “code_1.0.R”

“Table SI2.csv” Data generated for Table SI-2 in the supporting information. Generated in “PAPER TABLES” code section of “code_1.0.R”

“FigureX_xxx.csv” Data generated for Figures in the manuscript. All generated in the “PLOT DATA TABLES” code section of “code_1.0.R”. Numbered according to the Figure.

Figures

r list.files(path="figures")

##  [1] "Figure1_BC.pdf"            "Figure1_BC.png"           
##  [3] "Figure2_map.pdf"           "Figure2_map.png"          
##  [5] "Figure2d_LandCover.png"    "Figure2e_BurnPct.png"     
##  [7] "Figure3_Budyko.pdf"        "Figure3_Budyko.png"       
##  [9] "Figure4_bars.pdf"          "Figure4_bars.png"         
## [11] "Figure5_QChangeMerge.pdf"  "Figure5_QChangeMerge.png" 
## [13] "Figure6_Recovery.pdf"      "Figure6_Recovery.png"     
## [15] "FigureSI1_RiceEmanuel.png" "FigureSI2.png"            
## [17] "FigureSI3.png"             "FigureSI4_bars.png"       
## [19] "FigureSI5_flowcomp.png"    "FigureSI6.png"            
## [21] "FigureSI7_Pbars.png"       "Merge_Figures1-6.pdf"

This folder contains all figures generated using “code_1.0.R” and shown in the manuscript and supporting information.

QGISproj

This folder contains the QGIS project file and extra geospatial data used to construct Figure 2 (site map).

• The file “Old_fire.jpg” is licensed under the Creative Commons Attribution-Share Alike 1.0 Generic license: Schumaker, D. (2003). Old Fire burning in the San Bernardino Mountains. Looking west from Strawberry Peak. Retrieved from https://commons.wikimedia.org/wiki/File:Old_fire.jpg

• The file “California.A2003298.2105.250m.jpg” is from: NASA. (2003, October 25). Grand Prix Fire and Piru Fire near Los Angeles. Greenbelt, MD, USA: NASA Goddard Space Flight Center, EOS Project Science Office, Visible Earth Catalog. Retrieved from https://visibleearth.nasa.gov/view.php?id=69159

• The shapefile “CA_shp” is extracted from the spData package in R. Roger Bivand, Jakub Nowosad and Robin Lovelace (2020). spData: Datasets for Spatial Analysis. R package version 0.3.3. https://CRAN.R-project.org/package=spData

• The shapefile “2019Budyko_LA_cities_subset” is extracted from the Natural Earth world cities dataset: Made with Natural Earth. Free vector and raster map data @ naturalearthdata.com.