Assessing impacts of wildfire on water and habitat quality in downstream aquatic ecosystems using remote sensing
This project focuses on the downstream effects of wildfires on lakes, reservoirs, and coastal water within the larger context of the USGS project on Post-fire Hazards and Impacts to Resources and Ecosystems (PHIRE): Support for Response, Recovery, and Mitigation. The overall objective of this work is to leverage the historic satellite remote sensing record to provide insight on the impact of wildfires on downstream water quality parameters. Wildfires decrease infiltration capacity and groundwater recharge as well as the flood attenuation capacity of vegetation, thereby increasing overland flow. Consequently, post-fire precipitation events mobilize wildfire-altered materials and transport them into streams and rivers within burned areas, which in turn drain into larger systems. This can cause large changes in surface water quality due to increased fluxes of ash, sediments, nutrients, carbon, and metals. Satellite-based remote sensing is well-suited to monitor these changes in water quality parameters, particularly in water bodies that are not well- instrumented.
Remote sensing imagery acquired by the MultiSpectral Instruments (MSI) carried by two European Space Agency (ESA) satellites, Sentinel 2 A & Sentinel 2 B (S2 A/B), is the primary data source for this project. The data collected by the S2 A/B MSI provide landscape-scale imagery every 5 days with a pixel size of 20m. The historic record of S2A/B MSI extends to Nov 2015 – present. This time period covers the two fires of interest for this study: The Dixie Fire and the Caldor Fire. Both fires occurred in 2021 in California and burned extensive areas in the Sierras. These fires burned portions of the watersheds of three reservoirs that provide drinking water and recreation: Lake Almanor, Lake Oroville, and Folsom Lake. Pardee Reservoir is also included in the area of interest as a reservoir with a watershed that has not been impacted by wildfire since 2015. Lakes Almanor, Oroville, and Folsom have different fire histories in the last 5 years. Lake Oroville was impacted by the Camp Fire (2018) and the North Complex Fire (2020). Lake Almanor and Lake Oroville were impacted by the Dixie Fire (2021). Folsom Lake was impacted by the Caldor Fire (2021) and the Mosquito Fire (2022). In addition, these three reservoirs have differing occurrences of Harmful Algal Blooms (HABs).
The goal of this study is an improved methodology for assessing effects of wildfires on water quality and aquatic habitat quality based on remote sensing. We will use the timeseries of S2 A/B derived water quality parameters to investigate the effects wildfire by comparing water quality in impacted and control sites in the months following the wildfire event, after the first major storm, after spring runoff, and a year post-fire, in order to answer the following questions:
- Does wildfire extent impact changes in turbidity and dissolved organic matter (DOM) in the following lakes: Folsom, Oroville, Almanor, and Pardee?
- Does wildfire severity impact changes in turbidity and DOM in the following lakes: Folsom, Oroville, Almanor, and Pardee?
- What is the timing of water quality changes?
- What is the impact of repeated fire occurrence?
- What is the frequency of HAB occurrence? What is the extent of HABs? What is the timing of HABs?
Remote sensing imagery
The full archive of S2 A/B imagery (2016 – present, 754 images) was downloaded from ESA’s Copernicus Dataspace Ecosystem for each of the reservoirs included in this study. This data will be atmospherically corrected to remote sensing reflectance values, and retrieval models for fDOM and turbidity will be applied to the imagery. The pixel values for those retrievals will be compiled into a timeseries of turbidity and DOM measurements for each reservoir, as well used to identify HAB occurrence. This dataset forms the basis for all subsequent analysis.
Field Campaign
A field campaign was undertaken in August 2023 to collect in situ samples at each reservoir. The dates of the field sampling were chosen to coincide with the overpass dates of the S2 A/B satellites. These data will be used to validate the remote sensing turbidity & DOM retrievals, as well as provide insight on reservoir conditions approximately 2 years after the fires occurred.
To collect discrete water samples, a submersible pump was lowered to approximately 0.5 m below the water surface. Unfiltered water samples were collected for lab analysis of dissolved organic carbon (DOC), absorbance and fluorescence of dissolved organic matter (DOM; samples also known as “optics”), particulate absorbance (Ap), chlorophyll a, particulate carbon (PC), particulate nitrogen (PN), and for benchtop analysis with a Seabird ac-s spectral absorption and attenuation instrument. Unfiltered samples for DOC, optics, chlorophyll a, and the ac-s were passed directly from the tubing fed by the submersible pump into sample bottles. Unfiltered samples for Ap, PC, and PN, were collected as grab samples from within 0.3 m of the water surface. Filtered samples were collected for determination of NH4, NO2, NO3, PO4, and total dissolved nitrogen (TDN) using a 0.45 µm Pall capsule filter fed by the submersible pump.
In addition, instrumentation was utilized to measure field parameters. Water temperature, specific conductance, dissolved oxygen, pH, turbidity, fluorescence of chlorophyll (fCHL), and fluorescence of dissolved organic matter (fDOM), were measured and recorded using an EXO2 multiparameter water quality sonde. Profiles of photosynthetically active radiation (PAR) were measured using two LiCor Li 192 quantum sensors, with one pointing up and the other pointing down. Profiles were measured until a) the sensors hit the lake bottom, b) extinction was reached, or c) there was no more cable to lower the instrument. Above water hyperspectral radiometry data were collected using an SVC 512i spectroradiometer.
Funding
Funding for this project is provided by the Robert T. Stafford Disaster Relief and Emergency Assistance Act (42 U.S.C. 5121 et seq.) and supplemental funding acts for Federal disaster relief activities. Through this funding USGS supports recovery efforts in declared natural disaster areas, to aid recovery efforts from widespread wildfires, devastating hurricanes, prolonged volcanic eruptions, and damaging earthquakes. This enables USGS to repair and replace equipment and facilities, collect high-resolution elevation data, and conduct scientific studies and assessments to support recovery and rebuilding decisions.
This project focuses on the downstream effects of wildfires on lakes, reservoirs, and coastal water within the larger context of the USGS project on Post-fire Hazards and Impacts to Resources and Ecosystems (PHIRE): Support for Response, Recovery, and Mitigation. The overall objective of this work is to leverage the historic satellite remote sensing record to provide insight on the impact of wildfires on downstream water quality parameters. Wildfires decrease infiltration capacity and groundwater recharge as well as the flood attenuation capacity of vegetation, thereby increasing overland flow. Consequently, post-fire precipitation events mobilize wildfire-altered materials and transport them into streams and rivers within burned areas, which in turn drain into larger systems. This can cause large changes in surface water quality due to increased fluxes of ash, sediments, nutrients, carbon, and metals. Satellite-based remote sensing is well-suited to monitor these changes in water quality parameters, particularly in water bodies that are not well- instrumented.
Remote sensing imagery acquired by the MultiSpectral Instruments (MSI) carried by two European Space Agency (ESA) satellites, Sentinel 2 A & Sentinel 2 B (S2 A/B), is the primary data source for this project. The data collected by the S2 A/B MSI provide landscape-scale imagery every 5 days with a pixel size of 20m. The historic record of S2A/B MSI extends to Nov 2015 – present. This time period covers the two fires of interest for this study: The Dixie Fire and the Caldor Fire. Both fires occurred in 2021 in California and burned extensive areas in the Sierras. These fires burned portions of the watersheds of three reservoirs that provide drinking water and recreation: Lake Almanor, Lake Oroville, and Folsom Lake. Pardee Reservoir is also included in the area of interest as a reservoir with a watershed that has not been impacted by wildfire since 2015. Lakes Almanor, Oroville, and Folsom have different fire histories in the last 5 years. Lake Oroville was impacted by the Camp Fire (2018) and the North Complex Fire (2020). Lake Almanor and Lake Oroville were impacted by the Dixie Fire (2021). Folsom Lake was impacted by the Caldor Fire (2021) and the Mosquito Fire (2022). In addition, these three reservoirs have differing occurrences of Harmful Algal Blooms (HABs).
The goal of this study is an improved methodology for assessing effects of wildfires on water quality and aquatic habitat quality based on remote sensing. We will use the timeseries of S2 A/B derived water quality parameters to investigate the effects wildfire by comparing water quality in impacted and control sites in the months following the wildfire event, after the first major storm, after spring runoff, and a year post-fire, in order to answer the following questions:
- Does wildfire extent impact changes in turbidity and dissolved organic matter (DOM) in the following lakes: Folsom, Oroville, Almanor, and Pardee?
- Does wildfire severity impact changes in turbidity and DOM in the following lakes: Folsom, Oroville, Almanor, and Pardee?
- What is the timing of water quality changes?
- What is the impact of repeated fire occurrence?
- What is the frequency of HAB occurrence? What is the extent of HABs? What is the timing of HABs?
Remote sensing imagery
The full archive of S2 A/B imagery (2016 – present, 754 images) was downloaded from ESA’s Copernicus Dataspace Ecosystem for each of the reservoirs included in this study. This data will be atmospherically corrected to remote sensing reflectance values, and retrieval models for fDOM and turbidity will be applied to the imagery. The pixel values for those retrievals will be compiled into a timeseries of turbidity and DOM measurements for each reservoir, as well used to identify HAB occurrence. This dataset forms the basis for all subsequent analysis.
Field Campaign
A field campaign was undertaken in August 2023 to collect in situ samples at each reservoir. The dates of the field sampling were chosen to coincide with the overpass dates of the S2 A/B satellites. These data will be used to validate the remote sensing turbidity & DOM retrievals, as well as provide insight on reservoir conditions approximately 2 years after the fires occurred.
To collect discrete water samples, a submersible pump was lowered to approximately 0.5 m below the water surface. Unfiltered water samples were collected for lab analysis of dissolved organic carbon (DOC), absorbance and fluorescence of dissolved organic matter (DOM; samples also known as “optics”), particulate absorbance (Ap), chlorophyll a, particulate carbon (PC), particulate nitrogen (PN), and for benchtop analysis with a Seabird ac-s spectral absorption and attenuation instrument. Unfiltered samples for DOC, optics, chlorophyll a, and the ac-s were passed directly from the tubing fed by the submersible pump into sample bottles. Unfiltered samples for Ap, PC, and PN, were collected as grab samples from within 0.3 m of the water surface. Filtered samples were collected for determination of NH4, NO2, NO3, PO4, and total dissolved nitrogen (TDN) using a 0.45 µm Pall capsule filter fed by the submersible pump.
In addition, instrumentation was utilized to measure field parameters. Water temperature, specific conductance, dissolved oxygen, pH, turbidity, fluorescence of chlorophyll (fCHL), and fluorescence of dissolved organic matter (fDOM), were measured and recorded using an EXO2 multiparameter water quality sonde. Profiles of photosynthetically active radiation (PAR) were measured using two LiCor Li 192 quantum sensors, with one pointing up and the other pointing down. Profiles were measured until a) the sensors hit the lake bottom, b) extinction was reached, or c) there was no more cable to lower the instrument. Above water hyperspectral radiometry data were collected using an SVC 512i spectroradiometer.
Funding
Funding for this project is provided by the Robert T. Stafford Disaster Relief and Emergency Assistance Act (42 U.S.C. 5121 et seq.) and supplemental funding acts for Federal disaster relief activities. Through this funding USGS supports recovery efforts in declared natural disaster areas, to aid recovery efforts from widespread wildfires, devastating hurricanes, prolonged volcanic eruptions, and damaging earthquakes. This enables USGS to repair and replace equipment and facilities, collect high-resolution elevation data, and conduct scientific studies and assessments to support recovery and rebuilding decisions.