If you want a simple download, click on Releases. There's also a version on the ORNL DAAC.

srdb

Global soil respiration database Last data update: 03 May 2025

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 scientific analysis.

Files in this repository

File | Description ----------------------------|------------------------------------------------ calculations | Directory of spreadsheets (generally data estimated from figures) expandedfieldnotes.txt | More info on some fields and their uses LICENSE | The MIT license: short, simple, and permissive R | R check script directory README.md | Generates this README srdb-datafields.txt | Metadata for the srdb-data file srdb-data.csv | Main database srdb-equationsfields.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 -------- | ------------------------------------------------------------ 20250503 | Data up to 2022 entered by Ava Cook and Jennifer Mendez Gomez. 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. 20200424 | A variety of new data from the Russian-language literature; thanks to Nazar Kholod. 20200409 | Ingest of LOTS of new data (through PY2017; srdb-data.csv goes from 7318 to 10311 rows) entered by @jmanzon, @darlinp, and @jinshijian. 20200220 | Removed fields Rs_max, Rs_maxday, Rs_min, and Rs_minday from database; they haven't been used and are very inconsistent. Added Collar_height and merged a bunch of new data for that fields as well as Collar_depth, Chamber_area, and Time_of_day. 20191103 | New fields (Collar_depth, Chamber_area, Time_of_day); first data from interns Darlin P. and Jason M.; data corrections by @jinshijian 20190915 | Added lots of new data entered by @rbcmrchs and @jinshijian; corrections to older data; broke equations columns into separate, new srdb-equations.csv file. 20190306 | Many corrections to longitude and latitude data; see issue #22. Thanks to @jinshijian. 20181101 | Fixed a number of bad sand:silt:clay values, duplicate records, and other problems. Thanks to @ValentineHerr of SI, and Shafer Powell and Debjani Deb of ORNL DAAC. 20180724 | Fixed a number of lat/lon problems, and reverted to purely decimal format for those fields. Thanks to Debjani Deb. 20180216 | Updated the R QC script so it works correctly with new dataset. 20180215 | All data through 2013 along with some 2014 and 2015. Fixes to partitioning (Rh_annual, Ra_annual) in some previous records (studies 2037, 5545, 5727, 6010, 6219). Study_midyear field changed so that X.5 is consistently middle of year X. Latitude and Longitude fields may now be either decimal or d-m-s format. Thanks to Mercedes Horn. 20150826 | Fixes to faulty Longitude and Latitude values and site names; thanks to Yaxing Wei of ORNL. Some 2013 data added. 20140827 | Shifted from Google Code (svn) to Github (git). No data changes. 20131218 | Pubs from 2012 added (466 new records). Two new fields (measurement interval and annual coverage); renamed CO2_method to Meas_method; many corrections to older data; new R script for error-checking and mapping; removed kmz file. 20120510 | Pubs from 2011 added, many other corrections to temp models. Figshare doi: 10.6084/m9.figshare.868954. 20110524 | A few late-breaking studies added, and kmz file updated. 20110513 | Many missing 2010 studies added, and a number of errors fixed. Thanks to Dan Metcalfe and Les Hook. 20110224 | More 2010 pubs added; three fields deleted: Chamber_method, CH4_flux, N2O_flux. These were all inconsistent or almost never used. 20101029 | Pubs from first half of 2010 added. 20100825 | A number of Age_disturbance fields corrected and filled in. 20100517 | PY2009 data added. Field reordered to match Biogeosciences ms. 20100222 | Partition_method field fixed for many records. Thanks to Myroslava Khomik.

How to use these data

Read the documentation! There are a few fields to be especially careful of, especially "Qualityflag", "CO2method" (e.g. many analyses will want to exclude soda lime measurements), "Manipulation" (you may want to filter to "None" to look at un-manipulated systems), "Ecosystemstate", and "Ecosystemtype".

How to get a publication

If you'd like to obtain a PDF of one of the database studies, open an issue or email Ben Bond-Lamberty, the maintainer. In either case please specify the study number.

How to cite these data

This data set is open, and no co-authorship is required if you make use of it. The relevant citation is:

• Jian, Vargas, Anderson-Teixeira, Stell, Herrmann, Horn, Kholod, Manzon, Marchesi, Paredes, and Bond-Lamberty (2021). A restructured and updated global soil respiration database (SRDB-V5), Earth Syst. Sci. Data 13:255–267, 2021, doi: 10.5194/essd-13-255-2021.

The older, original citation is:

• Bond-Lamberty and Thomson (2010). A global database of soil respiration measurements, Biogeosciences 7:1321-1344, doi: 10.5194/bgd-7-1321-2010.

Analyses using these data should include the database version (i.e. the tag, such as "v20131218a", and ideally the git commit hash), download date, and URL. (At some point in the future, releases will include a DOI, too.)

How to contribute

The goal is for this to be a dynamic, community database, not just an archived blob. Why? Well, these data undoubtedly have mistakes and omissions; the database could be structured more usefully for new analyses; and the dataset should be extended as new studies are published. Internet-based hosting services like give us the ability to build a 21st-century data set that is modified by, and grows with, the needs of the scientific community.

There are at least ways to contribute. The first, and preferred, way is to fork out (using git) a copy of the data, just as if this were an open-source software project. You can then modify or add data and send me a pull request. The second way is to open an issue.