Major catchup (#142)

* Data, studies, code tweaks for 2021-2022
* Add misc calculation files
* Fix bad USA longitudes
* Fix a LOT of duplicated studies
* Update srdb_check.R, README.md, and check-srdb.yaml

.github/workflows/check-srdb.yaml

@@ -16,11 +16,11 @@ jobs:
     env:
       GITHUB_PAT: ${{ secrets.GITHUB_TOKEN }}
     steps:
-      - uses: actions/checkout@v2
+      - uses: actions/checkout@v3
         with:
           fetch-depth: 2
-      - uses: r-lib/actions/setup-r@v1
-      - uses: r-lib/actions/setup-pandoc@v1
+      - uses: r-lib/actions/setup-r@v2
+      - uses: r-lib/actions/setup-pandoc@v2
       - name: Install any needed packages
         run: |
           install.packages(c("ncdf4"))

R/srdb_check.R

@@ -1,9 +1,12 @@
-# Run checks on SRDB data for consistency, columns types, out of bounds, etc.
-# BBL 2019-09-17
+# Checks SRDB data for consistency, columns types, out of bounds, etc.,
+# BBL 2019-09-17; updated 2024-09-27
 
+if(basename(getwd()) != "srdb") stop("Working directory must be srdb/")
 
 # Helper functions ---------------------------------------------------------
 
+contiguous <- function(x) all(diff(x) == 1)
+
 check_numeric <- function(d, dname = deparse(substitute(d))) { 
   message("Checking ", dname, " is numeric")
   if(!is.numeric(d)) {
@@ -85,6 +88,16 @@ stop("Study_number values in srdb-data not found in srdb_studies: ",
      paste(unique(srdb$Study_number[!present]), collapse = ", "))
 }
 
+# Every study number should be contiguous in the data file
+contig_problem <- FALSE
+for(st in unique(srdb$Study_number)) {
+  x <- which(srdb$Study_number == st)
+  if(!contiguous(x)) {
+    message("\tStudy ", st, " is not contiguous")
+    contig_problem <- TRUE
+  }
+}
+if(contig_problem) stop("Non-contiguous study numbers are probably duplicates")
 
 # srdb-info.txt ---------------------------------------------------------
 
@@ -120,12 +133,29 @@ with(srdb, {
 	check_bounds(YearsOfData, c(1, 99))
 	check_bounds(Latitude, c(-90, 90))
 	check_bounds(Longitude, c(-180, 180))
-	check_bounds(Elevation, c(0, 7999))
+
+	# Special check for U.S. studies with mis-entered longitude
+	bad_USA <- srdb$Country == "USA" & srdb$Longitude > 0 & srdb$Region != "Guam"
+	bad_USA[is.na(bad_USA)] <- FALSE
+	if(any(bad_USA)) {
+	  stop("'USA' entries with positive longitude: ", 
+	       paste(which(bad_USA), collapse = " "))    
+	}
+
+	check_bounds(Elevation, c(-10, 7999))
+
+	# Manipulation is a mess, unfortunately
+	controls <- srdb$Manipulation %in% c("Control", "CK", "CT")
+	if(any(controls)) {
+	  stop("'Control' or 'CK' or 'CT' entries in Manipulation; should be 'None': ", 
+	       paste(which(controls), collapse = " "))    
+	}
+
 	check_bounds(Age_ecosystem, c(0, 999))
 	check_bounds(Age_disturbance, c(0, 999))
 	check_labels(Ecosystem_state, c("Managed", "Unmanaged", "Natural", ""))
 
-	# Ecosystem_type is really varied, unfortunately
+	# Ecosystem_type is also varied
 	unmanaged_ag <- srdb$Ecosystem_type == "Agriculture" & srdb$Ecosystem_state != "Managed"
 	if(any(unmanaged_ag)) {
 		stop("Non-managed agriculture in rows: ", paste(which(unmanaged_ag), collapse = " "))    
@@ -153,10 +183,10 @@ with(srdb, {
 	check_bounds(Soil_sand, c(0.0, 999.9))
 	check_bounds(Soil_silt, c(0.0, 999.9))
 	check_bounds(Soil_clay, c(0.0, 999.9))
-	check_bounds(MAT, c(-30, 40))
+	check_bounds(MAT, c(-30, 55))
 	check_bounds(MAP, c(0, 9999))
 	check_bounds(PET, c(0, 9999))
-	check_bounds(Study_temp, c(-30, 40))
+	check_bounds(Study_temp, c(-30, 46))
 	check_bounds(Study_precip, c(0, 9999))
 	# TODO: meas_method one of a few values
 	check_bounds(Meas_interval, c(0.01, 365))

README.md

@@ -4,36 +4,39 @@
 # srdb
 
 Global soil respiration database
-Last data update: 09 October 2022
+Last data update: 12 October 2024
 
 ## About
 
 The global soil respiration database (SRDB) is a database of published
 studies about the flux of CO2 from the soil surface to the atmosphere ("soil 
-respiration"") in the field. It's intended to serve as a resource for
+respiration") in the field. It's intended to serve as a resource for
 scientific analysis.
 
 ## Files in this repository
 
 File						|	Description
 ----------------------------|------------------------------------------------
 calculations				|	Directory of spreadsheets (generally data estimated from figures)
-corrections-additions.xls	|	Excel workbook for corrections/additions 
 expanded_field_notes.txt	|	More info on some fields and their uses 
-LICENSE | The SRDB uses the MIT license, which is short, simple, and permissive
+LICENSE                     |   The MIT license: short, simple, and permissive
 R							|	R check script directory
 README.md					|	Generates this README
 srdb-data_fields.txt		|	Metadata for the srdb-data file
 srdb-data.csv				|	Main database
-srdb-equations_fields.csv	|	Metadata for the srdb-equations file_
+srdb-equations_fields.csv	|	Metadata for the srdb-equations file
 srdb-equations.csv			|	Temperature and moisture sensitivity equations
 srdb-studies_fields.txt		|	Metadata for the srdb-studies file
 srdb-studies.csv			|	Studies database
 
+Note that the `srdb-equations.csv` entries used to be part of the main data file,
+but it was a lot of work, and didn't seem to be used, so we stopped updating it.
+
 ## Major changelog
 
 Date	   	|	Change
 -------- | ------------------------------------------------------------
+20241012 | Data up to 2022 entered by Gabriela Aguilar-Martinez, Melat Ghebreselassie, Alanna Hart, Brandon Kim, and Carly Pierce.
 20221009 | New 2018-2019 data entered by Elon Atlaw, Mahlet Dagnachew, and Emily Kang. 
 20210621 | Many (400+) corrections to location, time, ecosystem type, etc.; thanks to Ni Huang! See PR #126.
 20210522 | New 2017 data entered by Shoshi Hornum and Hope Ng.