T
T
T

Caspar Creek Experimental Watershed

Caspar Creek North Fork Weir, Jackson State Demonstration Forest
Photo Credit
U.S. Forest Service photo

This v-notch weir on the North Fork of Caspar Creek is used to measure streamflow

Located on the Jackson Demonstration State Forest near Fort Bragg, California, the North Fork and South Fork tributary watersheds of Caspar Creek serve as research sites for evaluating the effects of timber management on streamflow, erosion and sedimentation. The Caspar Creek Experimental Watersheds serve as an important long-term research site for providing information needed to manage forests in a manner that minimizes impacts on water quality and other resources. One of the key outcomes from Caspar Creek has been increasing our basic understanding of how forests, water, soil, and sediment interact under forest management.   Caspar Creek research evaluates the effects of timber management on streamflow, soil erosion, and sedimentation in the rainfall-dominated redwood forest watersheds of the northern coast of California. 

Since 1961, the USDA Forest Service Pacific Southwest Research Station has partnered with the California Department of Forestry and Fire Protection (CAL FIRE) to understand the effects of timber harvest and forest road management on watershed processes and outputs. The long record as well as new hillslope or watershed scale experiments allow Caspar Creek to answer current questions, extend our basic knowledge, and to guide land management practices.

Background Information and History

Storms of the 1950s hit hard in northwest California forests, triggering landslides that muddied streams and altered freshwater habitat for Pacific salmon. To what extent did forest land-use activities contribute to these and other water quality impacts? We didn’t know, and to find out we would need to measure the amount of sediment produced by logging-related activities. The Forest Service and the California Department of Forestry and Fire Protection jointly decided to study the effects of logging on the quality and quantity of water flowing from coastal watersheds. 

The Caspar Creek Experimental Watersheds are located on the Jackson Demonstration State Forest near Fort Bragg, California. The Caspar Creek Study was established in 1961 as a cooperative effort between the California Department of Forestry and Fire Protection (CALFIRE) and the Forest Service's Pacific Southwest Research Station. Since then, it has evolved from a simple paired watershed study into one of the most comprehensive and detailed investigations of its kind. Two forks of Caspar Creek—the North Fork (NFC, 479 ha or 1184 acres) and South Fork (SFC, 417 ha or 1030 acres)—plus twenty contributing sub-watersheds are currently equipped with instruments to measure streamflow and sediment. The 60+ year record from the Caspar Creek is unique in providing long-term hydrologic data from small, rain-dominated watersheds located in second and third growth conifer forests. Recognizing the value of this work, PSW and CAL FIRE have a 100-year Memorandum of Understanding to continue research at Caspar Creek at least through 2099.

Land Use Acknowledgement 

Caspar Creek is located on the traditional territory of the Northern Pomo Tribes. We acknowledge the Northern Pomo as the past, present, and future caretakers of the land where our research is centered.

Ecology

Climate

Winters are mild and wet, and summers are moderately cool and dry. About 90 percent of the average annual precipitation of 1,200 mm falls during October through April. Summer coastal fog is common and contributes to the water available for use by vegetation. Snow is rare and rainfall intensities are low.

Soils

The soils of the watersheds are well-drained clay-loams, 1 to 2 m in depth, and are derived from Franciscan graywacke sandstone and weathered, coarse-grained shale of Cretaceous Age. They have high hydraulic conductivities, and subsurface stormflow is rapid, producing saturated areas for brief periods over small areas.

Vegetation

The old-growth redwood forest at Caspar Creek was removed during large-scale logging across the region in the late 19th century. Current forest types include second- and third-growth mixed conifer, mostly coast redwood, Douglas-fir, western hemlock, and grand fir. Under the tree canopy, vegetation includes smaller trees like alder and tanoak as well as shrubs and forbs like evergreen huckleberry and ferns.

Facilities Information

In addition to the infrastructure within Caspar Creek, there is a small office and workstation located at the Jackson Demonstration State Forest office, 802 N. Main Street, Fort Bragg, CA. Short term lodging for field staff and research partners is available at the nearby Forest Learning Center. Additional staff offices and laboratory facilities are located at the Redwood Sciences Laboratory on the edge of the Cal Poly Humboldt campus in Arcata, CA.

Contact

Research and Findings

Research at Caspar Creek is designed to study the effects of forest practices on watershed and ecological processes. The main experimental design of monitoring two or more watersheds or sub-watersheds for a “calibration” period, then monitoring the watersheds for a “treatment” period after some management activity provides a rigorous approach to understanding the effects of the treatment. This before-after control-impact approach has been used throughout the history of the Caspar Creek Experimental Watersheds. Below is a timeline of experimental harvests and major findings.

Calibration: 1963 to 1967

The North Fork (NFC) and South Fork (SFC) watersheds were calibrated prior to treatment. At that time, these watersheds supported a 90-year-old, second-growth, mixed conifer forest dominated by coast redwood and Douglas-fir. 

First Experimental Harvest: 1967 to 1973

This experiment was carried out prior to implementation of modern California Forest Practice Rules. Roads were constructed and about two-thirds of the stand volume was harvested using selection silvicultural techniques and ground based transport of logs from SFC. NFC served as the experimental control and had no road construction or harvest during the calibration or treatment period.

Findings: The logging quadrupled suspended sediment yields in the South Fork . During the late 1960’s, following road construction in the watershed, salmonid abundance declined but appeared to return to pre-disturbance levels after only 2 years. Sediment loads declined to near pre-treatment levels by 1985. However, loads again increased in the early 1990s due to deterioration of the old road network, and these delayed sediment inputs are rivaled the volume introduced during the initial logging period. Monitoring also showed increased peak flows and low flows immediately after logging. Peak flows soon declined to pre-treatment levels, but low flows also dropped below pre-treatment levels until the early 2000s because of the high water use required for forest regrowth. 

Second Experimental Harvest: 1989 to 1991

The second experiment was carried out under the California Forest Practice Rules in effect at that time.  In 1985, the NFC was divided into ten smaller tributary sub-watersheds, and the calibration period monitoring began. Seven of the sub-watersheds were partially or fully clearcut harvested using cable-yarding techniques and three sub-watersheds were left unharvested as untreated experimental controls. Minimal new roads were required to implement the timber harvests. In addition to the experimental harvest, 67 percent of an 87-ha ungauged tributary was clearcut and cable yarded immediately upstream of the North Fork gauging station in 1985-1986.  

Findings: Results showed that sediment increases were lower in NFC during the second experiment than those from SFC in the first experiment, though increases were still observed. Annual suspended sediment loads increased by 73 percent in the partially clearcut NFC and by more than 100 percent in the clearcut tributary watersheds. Flow conditions and sediment loads had re-approached pretreatment levels by 2000. Like the first harvest, sediment yields later increased after their initial decline. Pre-commercial thinning in 2001 reactivated sediment sources, and a major landslide in 2006 from a 1990 clearcut contributed a considerable volume of sediment. 

The amount of large wood in NFC increased in the first few years after logging due to increased windthrow affecting the streamside buffers. Channel morphology has become more complex with more wood jams, sediment storage "steps," and greater pool volumes. However, no changes in the abundance of steelhead, coho, and Pacific giant salamander were attributed to logging. We continue to monitor some of the North Fork sub-watersheds to better understand long-term watershed recovery from timber harvest. 

Third Experimental Harvest: 2017 to 2019

The overarching objective of the third experiment is to better understand how reducing forest stand density within the limits of the current California Forest Practice Rules affects watershed processes. In 2010, ten sub-watersheds upstream of the South Fork weir were gauged and the calibration period began.  Stand density was reduced in six of the sub-watersheds between 20 and 76 percent by selection harvest and mostly cable yarding. Two of the sub-watersheds are controls and two others integrate the effects of harvest and no harvest at larger scales. 

Researchers are currently studying the effects of the third harvest on multiple aspects of the watershed’s hydrology and ecology, including hillslope hydrology, streamflow, erosion, sediment loads, road design, in-stream wood, wood decomposition, and macroinvertebrates. This research will result in a mechanistic understanding of the connection between forest density reduction and watershed processes that can be used to develop and continue to improve management practices in similar Coast Range watersheds as well as across the western U.S. 

Research Opportunities

Caspar Creek is a unique watershed observatory with substantial hydrological infrastructure at hillslope to watershed scales. We depend on collaboration and welcome new research at Caspar Creek. Please contact the lead scientist for more information about working at Caspar Creek. The Caspar Creek Experimental Watersheds also have research data archives available for public use. These long-term records are available through links in our data page.

Long-term Monitoring and Data

Long-Term Databases

There are long-term databases on stream discharge, precipitation, suspended sediment concentration, turbidity, and solar radiation.

Caspar Creek research results are commonly applied at least four ways:

  1. Results have been used in the development and refinement of California Forest Practice Rules that address watershed-related topics and have provided the information needed for state agency personnel to design strategies for controlling hydrologic changes from logging in coast redwood forests.
  2. Professional foresters, agency personnel, and landowners preparing management plans for forested areas need to describe potential effects of planned activities on water resources, and Caspar Creek results are often cited to support these estimates. Results have been used to predict changes in peak flows, hillslope erosion, sediment delivery, annual streamflow, low flows, and changes in stream temperature after logging.
  3. Caspar Creek data are available on the internet, and researchers from across the globe have used the data in support of their own research.
  4. Caspar Creek research has resulted in development of new methods for monitoring water quality, and the “Turbidity Threshold Sampling” monitoring strategy developed for efficient estimation of suspended sediment loads using turbidity measurements has now been adopted at sites around the world. 
     
Stream monitoring data from the South Fork Caspar Creek monitoring stations

Caspar Creek Data

Since 1961, records from the North and South Forks of Caspar Creek have provided long-term hydrologic data from small, rain-dominated watersheds located in second and third-growth conifer forests.

Types of Data

Hydrologic data include streamflow, suspended sediment concentrations and loads, turbidity, bedload sediment transport rates and loads, soil pipe flow, groundwater levels, soil moisture and tension, sapflow rates, air and water temperature, precipitation, water isotopes from various sources, and solar radiation.

Channel data include channel cross sections, stream habitat typing, large in-stream wood inventories, sediment storage, and pool condition.

Hillslope data include landslide inventories, road evaluations, and vegetative conditions.

Cooperators have studied benthic macroinvertebrates, aquatic vertebrates, water chemistry, and wood decomposition.

Accessing Data

Please contact one of our science staff to discuss opportunities for collaboration or for help accessing data.

We recommend using the following acknowledgements:

  • For use of Caspar Project data: “Data were provided by the Caspar Creek Experimental Watersheds project, which was funded by the USDA Forest Service Pacific Southwest Research Station and the California Department of Forestry and Fire Protection.”
  • And for use of Caspar site or facilities: “We conducted this research at the Caspar Creek Experimental Watersheds, which are located on the Jackson Demonstration State Forest and funded by the USDA Forest Service Pacific Southwest Research Station and the California Department of Forestry and Fire Protection.”

Data on recent conditions are hosted by Cal Poly Humboldt in cooperation with PSW

  • Plot current streamflow, sediment, rainfall, and temperature [Currently offline for maintenance]

Caspar Creek Experimental Watersheds Hydrologic and Climatic Data

Data include:

  • Streamflow & Sediment (1962 – 2017)
  • Precipitation (1985-2017) and meteorological conditions (2009 – 2017)
  • Geodatabase and shapefiles

Richardson, Paul W.; Seehafer, Jayme E.; Keppeler, Elizabeth T.; Sutherland, Diane G.; Wagenbrenner, Joseph W. 2021. Caspar Creek Experimental Watersheds Phase 1 (1962-1985) data. 2nd edition. Fort Collins, CO: Forest Service Research Data Archive. https://doi.org/10.2737/RDS-2020-0017-2

Richardson, Paul W.; Seehafer, Jayme E.; Keppeler, Elizabeth T.; Sutherland, Diane G.; Wagenbrenner, Joseph W.; Bladon, Kevin D.; Dymond, Salli F.; Cole, Ryan P.; Lininger, Katherine B.; Hilton, Susan J. 2022. Caspar Creek Experimental Watersheds Phase 2 (1985-2017) data. 3rd Edition. Updated 10 January 2024. Fort Collins, CO: Forest Service Research Data Archive. https://doi.org/10.2737/RDS-2020-0018-3

Additional data (e.g., tributary stations, meteorological conditions, GIS files, etc.) will be periodically added to these archives. Additional data may be available upon request to Joe Wagenbrenner.

Turbidity threshold sampling is an automated procedure for measuring turbidity and sampling suspended sediment. The basic equipment consists of a programmable data logger, a turbidimeter mounted in the stream, a pumping sampler, and a stage-measuring device. The data logger program uses turbidity to govern sample collection during each transport event.

Related Programs

People

  • Person

    Joseph Wagenbrenner, PhD

    Research Hydrologist
  • Person
  • Person
    USDA Forest Service Logo

    Anna Thompson

    Hydrologist
  • Person

    Diane Sutherland

    Geomorphologist
  • Person
  • Person

    Brian Storms

    Hydrologic Technician
  • Person

    Brian Eichel

    Hydrologic Technician
  • Person

    Susan J. Hilton

    Hydrologist

Collaborators

  • California Department of Forestry and Fire Protection

  • California Department of and Fish and Wildlife

  • California Water Quality Control Board

  • University of California at Davis

  • Cal Poly Humboldt

  • Cal Poly San Luis Obispo

  • Oregon State University

  • Northern Arizona University

  • University of Minnesota, Duluth

Multimedia

Related Programs

Publications available on Treesearch (alphabetical by title)

Last updated June 10, 2026