The data presented in this data release represent observations of postfire debris flows that have been collected from publicly available datasets. Data originate from 13 different countries: the United States, Australia, China, Italy, Greece, Portugal, Spain, the United Kingdom, Austria, Switzerland, Canada, South Korea, and Japan. The data are located in the file called “PFDF_database_sortedbyReference.txt” and a description of each column header can be found in both the file “column_headers.txt” and the metadata file (“Post-fire Debris-Flow Database (Literature Derived).xml”). The observations are derived from areas that have been burned by wildfire and are global in nature. However, this dataset is synthesized from information collected by many different researchers for different purposes, and therefore not all fields are available for each of the observations. Missing information is indicated by the value “-9999” in the ”PFDF_database_sortedbyReference.txt” file. Note that the text file contains special characters and a mix of date-time formats that reflect the original data provided by the authors. The text may not be displayed correctly if it is opened by proprietary software such as Microsoft Excel but will appear correctly when opened in a text editor software.

This table contains measured and modeled postfire debris flow volumes alongside the associated sources for debris flow documentation, locations, and volumes. We conducted a search of scientific literature and news media reports to find documentation of debris flows that may have followed all wildfires greater than 100 square kilometers that occurred between 1984 and 2021 in California. The wildfires listed are all the fires we found that had documented postfire debris flows. Some fires had field-measurements of debris flow volume. Where field-measurements of volume did not exist, we used model data on postfire debris-flow likelihood and volume from U.S. Geological Survey Emergency Assessment of Post-Fire Debris-Flow Hazards Team (see Lineage section in this metadata). In some cases, exact locations (but not volumes) were known, in which case we used modeled volumes for these locations. Where debris flows were documented but not exact locations, we used the volumes from all basins with a probability greater than eighty percent of having postfire debris flows.

This table contains measured and modeled postfire debris flow volumes alongside the associated sources for debris flow documentation, locations, and volumes. We conducted a search of scientific literature and news media reports to find documentation of debris flows that may have followed all wildfires greater than 100 square kilometers that occurred between 1984 and 2021 in California. The wildfires listed are all the fires we found that had documented postfire debris flows. Some fires had field-measurements of debris flow volume. Where field-measurements of volume did not exist, we used model data on postfire debris-flow likelihood and volume from U.S. Geological Survey Emergency Assessment of Post-Fire Debris-Flow Hazards Team (see Lineage section in this metadata). In some cases, exact locations (but not volumes) were known, in which case we used modeled volumes for these locations. Where debris flows were documented but not exact locations, we used the volumes from all basins with a probability greater than eighty percent of having postfire debris flows.

This data release is a field-verified inventory of postfire debris flows for the 2021 Dixie Fire following a 23-25 October 2021 atmospheric river storm and 12 June 2022 thunderstorm. The “README.txt” file describes the fields for the “Inventory.csv” file. The “Chambers” and “Chips” rain gage data referenced in the inventory are included as: “Chambers-Oct2021-Storm.csv”, “Chambers-Jun2022-Storm.csv”, “Chips-Oct2021-Storm.csv”, and “Chips-Jun2022-Storm.csv.” The fields for the rain gage data, which includes the geographic locations of the gages, are also described in the “README.txt” file. Fields with value “-9999” indicate that data are not available or do not exist.

This data release is a field-verified inventory of postfire debris flows for the 2021 Dixie Fire following a 23-25 October 2021 atmospheric river storm and 12 June 2022 thunderstorm. The “README.txt” file describes the fields for the “Inventory.csv” file. The “Chambers” and “Chips” rain gage data referenced in the inventory are included as: “Chambers-Oct2021-Storm.csv”, “Chambers-Jun2022-Storm.csv”, “Chips-Oct2021-Storm.csv”, and “Chips-Jun2022-Storm.csv.” The fields for the rain gage data, which includes the geographic locations of the gages, are also described in the “README.txt” file. Fields with value “-9999” indicate that data are not available or do not exist.

Summary This data release is an inventory of runoff-generated postfire debris flows compiled from 17 burn areas across five western U.S. states. Debris-flow data from the following fires are included: - Arizona: 2017 Pinal and 2019 Woodbury Fires - California: 2020 Apple, 2020 Bond, 2015 Butte, 2020 El Dorado, 2014 El Portal, 2018 Ferguson, 2016 Fish (San Gabriel Complex), 2011 Motor, and 2017 Thomas Fires - Colorado: 2020 Cameron Peak and 2018 Spring Creek Fires - New Mexico: 2018 Buzzard Fire - Washington: 2021 Cedar Creek, 2021 Cub Creek 2, and 2021 Muckamuck Fires The included table, “Combined_Inventory.csv”, contains debris-flow records represented as “1” or “0”, indicating whether a debris flow did occur or did not occur, respectively, in response to a given storm. In addition, for each observation, the corresponding fire name, observation location, storm date, and rainfall intensities are included. The “README.txt” file describes the fields (including units) for “Combined_Inventory.csv” in more detail. "0" values in the "Response" column indicate that no debris flow was observed, but a hydrologic response with lower sediment content may or may not have occurred. Assembling these data in a unified format involved a combination of compiling data from published sources and synthesizing new debris-flow records from observations made by US Geological Survey staff and partners. All records in the included file, "Combined_inventory.csv", are credited according to the individuals or publications that provided the observation information in the column labeled “ObservationSource”. To the best of our knowledge, all records included here correspond to debris flows generated by runoff, not by other mechanisms. These records have only limited spatial and temporal extent: they should not be taken as a comprehensive record of the hydrologic response in every catchment in response to every postfire storm. Fields with value “-9999” indicate that data are not available or do not exist. Data Credits Special thanks go to many colleagues who provided or helped to assemble debris-flow observation and/or rainfall data, including: Ann Youberg (Arizona Geological Survey); Paul Burgess, Jeremy Lancaster, Don Lindsay, John Oswald, and Brian Swanson (California Geological Survey); Chad Neptune (California State Water Resources Control Board); Stephanie Kampf, Lee MacDonald, Megan Sears, and Ellen Wohl (Colorado State University); Luke McGuire (University of Arizona); Maya Daurio (University of British Columbia); Laura Hempel, Jason Kean, Francis Rengers, Brittany Selander, and Matthew Thomas (U.S. Geological Survey); and Mitchell Allen and Kate Mickelson (Washington Geological Survey).

Summary This data release is an inventory of runoff-generated postfire debris flows compiled from 17 burn areas across five western U.S. states. Debris-flow data from the following fires are included: - Arizona: 2017 Pinal and 2019 Woodbury Fires - California: 2020 Apple, 2020 Bond, 2015 Butte, 2020 El Dorado, 2014 El Portal, 2018 Ferguson, 2016 Fish (San Gabriel Complex), 2011 Motor, and 2017 Thomas Fires - Colorado: 2020 Cameron Peak and 2018 Spring Creek Fires - New Mexico: 2018 Buzzard Fire - Washington: 2021 Cedar Creek, 2021 Cub Creek 2, and 2021 Muckamuck Fires The included table, “Combined_Inventory.csv”, contains debris-flow records represented as “1” or “0”, indicating whether a debris flow did occur or did not occur, respectively, in response to a given storm. In addition, for each observation, the corresponding fire name, observation location, storm date, and rainfall intensities are included. The “README.txt” file describes the fields (including units) for “Combined_Inventory.csv” in more detail. "0" values in the "Response" column indicate that no debris flow was observed, but a hydrologic response with lower sediment content may or may not have occurred. Assembling these data in a unified format involved a combination of compiling data from published sources and synthesizing new debris-flow records from observations made by US Geological Survey staff and partners. All records in the included file, "Combined_inventory.csv", are credited according to the individuals or publications that provided the observation information in the column labeled “ObservationSource”. To the best of our knowledge, all records included here correspond to debris flows generated by runoff, not by other mechanisms. These records have only limited spatial and temporal extent: they should not be taken as a comprehensive record of the hydrologic response in every catchment in response to every postfire storm. Fields with value “-9999” indicate that data are not available or do not exist. Data Credits Special thanks go to many colleagues who provided or helped to assemble debris-flow observation and/or rainfall data, including: Ann Youberg (Arizona Geological Survey); Paul Burgess, Jeremy Lancaster, Don Lindsay, John Oswald, and Brian Swanson (California Geological Survey); Chad Neptune (California State Water Resources Control Board); Stephanie Kampf, Lee MacDonald, Megan Sears, and Ellen Wohl (Colorado State University); Luke McGuire (University of Arizona); Maya Daurio (University of British Columbia); Laura Hempel, Jason Kean, Francis Rengers, Brittany Selander, and Matthew Thomas (U.S. Geological Survey); and Mitchell Allen and Kate Mickelson (Washington Geological Survey).

This data release includes time-series data from a monitoring site located in a small (0.12 km2) drainage basin in the Las Lomas watershed in Los Angeles County, CA, USA. The site was established after the 2016 Fish Fire and recorded a series debris flows in the first winter after the fire. The station is located along the channel at the outlet of the study area (34 9’18.50”N, 117 56’41.33”W, WGS84). The data were collected between November 15, 2016 and February 23, 2017. The data include two types of time series: (1) continuous 1-minute time series of rainfall and flow stage recorded by a laser distance meter suspended over the channel (LasLomasContinuous.csv), and (2) 50-Hz time series of flow stage and flow-induced ground vibrations recorded by two geophones (LasLomasStorm.csv). The continuous file contains brief data gaps when the station was serviced, at which time the record of cumulative rainfall was reset to zero. The ground vibrations were measured by two 4.5 Hz vertical axis geophones (Geospace SNG 11D/PC902/OPEN-30m) located approximately 2 m from the channel bank. One geophone was located 6.4 m upstream from the laser distance meter. The second geophone was located 7.6 m downstream of the geophone. The geophone data is recorded in millivolts and the geophone constant is 32 Volts/(m/s). The equation for converting the laser distance measurements into flow stage above the bedrock in the channel is: Stage_laser (meters) = 4.320 meters – Distance_laser (millimeters) /1000. Time stamps are in Coordinated Universal Time (UTC). Details of this study are described in the journal article: Tang, H., McGuire, L.A., F.K. Rengers, Kean, J. W., Staley, D.M., and Smith, J.B. (2018), Evolution of debris flow initation mechanisms and sediment sources during a series of post-wildfire rainstorms, J. Geophys. Res., xxx, FYYYYY, doi:10.1029/2018JF004837.

This data release includes time-series data from a monitoring site located in a small (0.12 km2) drainage basin in the Las Lomas watershed in Los Angeles County, CA, USA. The site was established after the 2016 Fish Fire and recorded a series debris flows in the first winter after the fire. The station is located along the channel at the outlet of the study area (34 9’18.50”N, 117 56’41.33”W, WGS84). The data were collected between November 15, 2016 and February 23, 2017. The data include two types of time series: (1) continuous 1-minute time series of rainfall and flow stage recorded by a laser distance meter suspended over the channel (LasLomasContinuous.csv), and (2) 50-Hz time series of flow stage and flow-induced ground vibrations recorded by two geophones (LasLomasStorm.csv). The continuous file contains brief data gaps when the station was serviced, at which time the record of cumulative rainfall was reset to zero. The ground vibrations were measured by two 4.5 Hz vertical axis geophones (Geospace SNG 11D/PC902/OPEN-30m) located approximately 2 m from the channel bank. One geophone was located 6.4 m upstream from the laser distance meter. The second geophone was located 7.6 m downstream of the geophone. The geophone data is recorded in millivolts and the geophone constant is 32 Volts/(m/s). The equation for converting the laser distance measurements into flow stage above the bedrock in the channel is: Stage_laser (meters) = 4.320 meters – Distance_laser (millimeters) /1000. Time stamps are in Coordinated Universal Time (UTC). Details of this study are described in the journal article: Tang, H., McGuire, L.A., F.K. Rengers, Kean, J. W., Staley, D.M., and Smith, J.B. (2018), Evolution of debris flow initation mechanisms and sediment sources during a series of post-wildfire rainstorms, J. Geophys. Res., xxx, FYYYYY, doi:10.1029/2018JF004837.

This data release is a geospatial record of flooding and debris-flow activity following the 2020 Dolan Fire along the central California coast. The postfire hydrologic responses documented here (“Inventory.csv”) are associated with a January 27, 2021, atmospheric river storm that hosted a narrow cold-frontal rain band, which produced intense rainfall over the burned area. Each entry in this record indicates whether no erosion, flooding activity, or debris-flow activity along a given channel reach was apparent in satellite-based imagery and/or during field-based mapping. The “README.txt” file describes the fields for all of the datasets. The “ProcessSteps.txt” file outlines how the data were collected. Additional details and analysis of this dataset are provided in Cavagnaro and others (2025). Reference Cited: Cavagnaro, D.B., McCoy, S.W., Thomas, M.A., Kostelnik, J., and Lindsay, D.N., 2025, Improved prediction of postfire debris flows through rainfall anomaly maps: Geophysical Research Letters, https://doi.org/10.1029/2025GL114791.