Bibliography Background About KRIS

KRIS Big River Info Links


Stream Temperature Monitoring by Jackson Demonstration State Forest and other Forest Science Project Cooperators 

The Jackson Demonstration State Forest (JDSF) comprises large portions of the Little North Fork, and North Fork sub-basin of the Big River, as well as Caspar Creek and other coastal watersheds in the area. JDSF began a water temperature monitoring program in 1996 that attempted to assess summer stream temperature at many locations using a limited number of automated temperature sensing probes. The Forest Science Project (FSP) in Arcata was contracted to organize JDSF temperature data and to remove outliers. FSP provided 1996-1998 water temperature data, and a spatial point coverage of monitoring site locations, to IFR for use in KRIS.

Water temperature references in KRIS Big River are extensively discussed in the Temperature background page.

Stream Temperature Monitoring by Mendocino County 

The Mendocino County Water Agency (MCWA) provided stream temperature data for two sites in the Big River for 1999, 2000, and 2001. The sites that were monitored were Lower Daugherty Creek and South Fork Big River below Daugherty Creek.

The MCWA uses Onset Hobo® Temp water temperature monitors, which can be expected to be accurate within 2.3ºF (1.3ºC). According to the user’s manual, this instrument will respond to temperature changes of between 0.7 to 1.6ºF (0.4 to 0.9ºC). Prior to placement in the streams, the temperature monitors were started and then immersed in an ice bath (0ºC) overnight. Although a National Institute of Standards and Technology (NIST) traceable thermometer is not used to check the temperatures during calibration, the response of this non-NIST thermometer has been checked against a NIST-traceable thermometer. Within 2 weeks, the temperature monitors are placed in the stream at the deepest locations with good shading. A deep, well shaded area is chosen due to an interest in refugia temperature. However, the temperature monitors are not necessarily placed in thermally well-mixed locations. Therefore, the water temperatures they record may vary slightly from average water temperature conditions in their respective thermal reaches.

Units were placed in pools by hand at a depth of about 1.5 - 2.5 feet (actual depths, deepest water available is selected) and are typically tied to a rock and then covered with rocks. If this is not possible, they are tethered to rebar that was driven into the streambed or placed close to the bank and hidden beneath vegetation to prevent vandalism. At the time of deployment, the stream temperature, air temperature, pool depth and time is recorded. Probes were set to collect temperature measurements every 96 minutes. Generally, a 96 minute collection interval is used to ensure capture of the daily maximum temperatures (Lewis et al 2000). During retrieval the water temperature, air temperature, pool depth and time are recorded. After retrieval, the data set is checked for proper response to the room and ice bath temperatures recorded prior to placing the temperature monitors in the field. This extraneous data is then trimmed from the data set prior to analysis.

Water temperature thresholds in KRIS Big River are extensively discussed in the Temperature Background page.

Stream Temperature Monitoring by Hawthorne Timber Company

Hawthorne Timber Company owns land in the Lower and Little North Fork sub-basins of the Big River. Stream Temperature monitoring of streams on those lands began in 1993 by Georgia-Pacific and is continued currently by Campbell Timberland Management. Campbell staff provided the KRIS project with 14 Excel files (one for each site) containing temperature data in the form of daily averages. Data for some sites was limited to a single year, or partial records for some years. Raw data was not provided. The summarized data also included a moving a 7DADM value which is the average of the 7 daily maximum values leading up to each date. This summary data was not used in favor of a floating weekly average (FWA) and floating weekly maximum (FWA) which includes the seven daily values around each date. Water temperature references in KRIS Big River are extensively discussed in the Temperature Background page. HTC water temperature monitoring site locations in the Big River are as follows:


Stream Temperature and Canopy Monitoring by Mendocino Redwood Company

The Mendocino Redwood Company (MRC) provided extensive water temperature data within the Lower, North Fork, Upper, South Fork sub-basins of the Big River for 1992-2001. Data included that collected by Louisiana Pacific Lumber Company during the 1992-1997 period before acquisition by MRC.

Data were collected using electronic temperature recorders (Stowaway, Onset Instruments) set to continuously monitor water temperature at 2 hour intervals. Monitoring occurred during the summer months when the water temperatures are highest, and recorders were typically placed in shallow pools (<2 ft. in depth) directly downstream of riffles. Data were received by the KRIS project twice, and always in the form of individual Excel files for each site-year dataset. Some files did not include raw data, but did include daily average and maximum values. MRC also provided an Arc View map coverage of temperature monitoring locations. A table of MRC temperature datasets is available for viewing the distribution of MRC data across space and time, as well as site codes and names. All datasets were included in KRIS Big River, using floating weekly average temperature. Water temperature thresholds in KRIS Big River are extensively discussed in the Temperature Background page. MRC also provided some stream canopy monitoring data for selected reaches in the Big River in 2000, with a description of canopy measurement methods and results.

Measurement Details:  Data were collected using Onset StowAway® continuous water temperature monitors. Prior to placement in the stream, each temperature monitor was calibrated with a 0ºC ice bath to ensure proper response to temperature. Given the type of water temperature monitors used, it is expected that the error would not exceed 0.23ºC. Monitoring occurred during the summer months when the water temperatures are highest, and recorders were typically placed in shallow pools (<2 ft. in depth) directly downstream of riffles. Data collection intervals varied from 72 to 144 minutes, but the majority of probes were set to collect temperature measurements every 120 minutes. Generally, a 96 minute collection interval is used to ensure capture of the daily maximum water temperatures (Lewis et al 2000). As the collection interval increases beyond 96 minutes, the peaks may be missed and it becomes increasingly likely that the recorded daily maximum water temperatures are lower than the actual daily maximum water temperatures.

Surface Water Ambient Monitoring Program

During 2001, Regional Water Board staff collected water quality measurements two times in the Big River watershed. This sample collection and analysis was done under the SWAMP program by SWAMP and NCWAP staff. While in the field, staff recorded basic "point in time" water chemistry that included pH, specific conductance, temperature, turbidity, and dissolved oxygen. Grab water samples were also collected and analyzed by an independent, certified laboratory. Water quality parameters that were analyzed included chlorophyll-a, alkalinity, hardness, various metals, and other compounds shown in the Source Table for these topics.

In total there were three (3) SWAMP sampling points in the Big River Watershed in 2001: the mainstem Big River just below the confluence of the Little North Fork of the Big River; the North Fork of the Big River, just below the confluence with Chamberlain Creek; and Daugherty Creek, just below the confluence of the South Fork of the Big River.

Sample collection (including field measurements) was done in accordance with the protocols described in the Regional Water Board appendix to the NCWAP Methods Manual (See Big River NCWAP report bibliography). Laboratory analysis was done in accordance with the appropriate US EPA analytical method.

US EPA Legacy StoRet

Extensive data on water quality was extracted from the US EPA legacy StoRet database (StoRet). As similar database maintained by USGS was found to only contain a subset of the data that was contained in StoRet, so the USGS database was not used.

The various water quality samples collected by USGS and the Regional Water Board were collected using accepted methods for the parameters during the time period in which they were collected. In some cases, collection and analysis methods may have changed substantially over the years. Depending on the parameter, data should be carefully evaluated to ensure that the data is of sufficient quality for the analysis (i.e. there is an adequate number of samples, the detection limit is appropriate, etc.). In those cases where data is found to be inadequate, follow up with a systematic sampling program using contemporary sampling and analysis techniques is advised.

California Department of Health Services

The water quality samples collected for the California DHS Community Well database were collected by the operators of the respective drinking water systems. While the source waters at the system intake is typically sampled, the location of the sample may vary. Depending on the actual sample location, water chemistry could vary significantly from the source waters. Also, the water quality data contained in the DHS database is generally collected on irregular intervals. For these reasons, the quality of the data is uncertain and therefore considered poor.

Water quality issues noted by the analysis of this data should be considered screening-level only. Any potential issues noted should be followed up with a systematic sampling program using US EPA approved sampling and analysis techniques.

California Department of Fish and Game Electrofishing Survey Data

The California Department of Fish and Game (CDFG) has contributed two sets of electrofishing data from Mendocino coastal area.  Weldon Jones and other CDFG staff began sampling various stream locations throughout Mendocino watersheds in 1983.  Since 1996, the effort has continued under the leadership of Scott Harris and generated several long term datasets, including those for Pudding Creek and Caspar Creek.  Electrofishing samples are generally taken in August through October when juvenile migration has abated and the resident fish population is most easily characterized.  The resulting database (CDFG biosample database) was shared with the KRIS project and used to demonstrate fish community structure for locations within the project area. Specific techniques used by CDFG staff, such as the use of blocking nets, and single versus multiple-passes varied. As a result, some of these data should not be used to compare fish abundance or fish densities over time or between reaches.  Information on sampling technique is currently limited to fields in the Source Table.  Presentation of the data in KRIS is intended to characterize the fish community structure for a variety of sites and years.  Chart and chart table captions describe some of the sampling details that effect how quantifiably each record may describe actual fish density or population trends.  Although the biosample database lists virtually all species of aquatic vertebrates that may occur, sample data indicates that amphibians and possibly some non-game fishes were not recorded until the most recent years.

The other set of electrofishing data from CDFG was collected as biological inventory during stream habitat inventories.  Habitat typing surveys conducted by CDFG between 1994-1999 involve crews staffed by California Conservation Corp and Americorp technicians with minimal experience.  Surveys follow methods described in the California Salmonid Restoration Manual (CDFG, 1998), and include electrofishing of representative habitats, but no use block nets, or any protocol to allow quantitative calculations. Sampling effort is minimal. Non detection of a species, such as coho salmon, cannot be taken as total absence.  Furthermore, these inventories are highly variable with respect to documenting non-game fishes and amphibians. Because of such limitations, these data are not highlighted in KRIS, but can be found in the stream inventory reports contained within the KRIS Bibliography.  Some examples of these data are used, however, as a simple indicator of fish species presence and fish community structure. The ratio of coho to steelhead, and the change in this ratio over time, reveal information about habitat change and suitability.

For more in-depth information see the background page on Fish Population and Fish Population Information in KRIS.

Fisheries Data from Hawthorne Timber Company (former Georgia Pacific)

Georgia-Pacific established two sites in the Big River basin as part of a monitoring program that began in 1993.  Such monitoring has been continued by Hawthorne Timber Company who supplied the data for use in KRIS.  These sites, one on the Little North Fork and another on Two Log Creek, were subject to annual electrofishing for the monitoring of aquatic vertebrates, as well as temperature monitoring and sediment monitoring.  For the purposes of electrofishing, the sites are actually 30 to 50 meter reaches that include a combination of pool and riffle habitat.  Reach limits are defined by change in habitat type, and 4.5 mm mesh blocking nets where used when sampling.  Sampling occurred in September or early October when flows are low and fish have become less mobile.  Data collected during the surveys include habitat area, number of vertebrates captured by species, and species biomass per site.  These data provide annual measures of abundance, density, biomass, and community structure which are useful in detecting trends in community structure and population levels.  Measured density and relative abundance are not accurate for all species due to different capture probabilities. Sculpin, for example, have low catch rates due to their ability to lodge themselves in the stream bottom.  Inference from this site-specific data to stream-wide conditions is may not be valid from a strict scientific perspective due to lack of replication and sample replication. Data from these sites do provide, however, valuable indices on the condition of juvenile salmonid and other stream fish populations.

Data from HTC electrofishing is presented in KRIS Big River for each site as either the total catch by species for a given, or time-series (1993-2000) for groups of amphibian or fish species. The total catch presentation utilizes a common KRIS chart style which is used for interpreting species composition or community structure.  Select years (one early in record and one late in record) have been presented as total catch topics.  Time-series data for each site is presented as biomass, density, and total catch of key species. Time-series data is presented for amphibian and fish species, separately, although non-represented species can be added to a graph by simply selecting an additional chart table column in the lowest open drop-down box to the right of the chart.

Fish Data from Mendocino Redwood Company

The Mendocino Redwood Company (MRC) provided both quantitative and non-quantitative electrofishing data for inclusion in KRIS Big River. Electrofishing can generate quantitative data when used block nets are placed above and below the area surveyed, and when multiple passes are made. Such data was collected for two sites in Gates Creek for the years 1990-1994, and for a site in the East Branch North Fork, and the Big River at Wildhorse Opening in the years 1993-1994. Only these data are available for investigating fish density, biomass, or changes in abundance. Another and larger set of fisheries data from MRC comes from single-pass electrofishing or snorkeler counts of many sites in the years 1994-1996, and 2000. The sites are distributed widely over stream courses on MRC land and surveyed for the purpose of detecting the presence of fish species. These data do not enable the assessment of fish health or abundance, but do provide a look at fish community structure, and specifically the presence of coho or other species. For more information see Fish Populations and Fish Population Information in KRIS.

Habitat Typing Surveys

Stream habitat surveys by the CDFG and the Georgia Pacific Company (now Hawthorne Timber Company) provide an inventory of stream conditions which is useful for assessing suitability for coho, and steelhead. Protocols follow the California Salmonid Restoration Manual (CDFG, 1998).   Of all data available from these surveys, four habitat attributes of particular importance (percent habitat type by length, maximum pool depth, embeddedness, and canopy) are charted for particular streams. The relative proportions of a stream falling under the simple habitat types of flatwater, riffle, and pools can be a useful indicator of fish habitat condition because coho salmon and 1+ steelhead require pool habitat for successful rearing. The deeper the pools the better and thus maximum pool depth is a useful measurement of habitat quality. Embeddedness from these surveys is used as a rough estimate of spawning gravel quality, although bulk gravel samples are far superior as a monitoring tool. Canopy measurements in habitat typing surveys are quantitative and give indications of opportunities for stream warming and long term prospects for large wood recruitment. In addition to these simple and useful charts, habitat data has also been presented in a basin wide format to facilitate analysis of the range and variability of habitat conditions both within and among streams. For habitat attributes, such as canopy, which vary greatly within a stream, mean values can be misleading or useless. As an alternative, box-whisker plots show the range and distribution of values. Box-whisker plots have been prepared in KRIS Big River to chart data for the four habitat attributes discussed above. In addition, shelter rating and residual pool volume have been charted. Shelter rating describes the complexity of habitat and can be a useful measure for assessing habitat quality. Residual pool volume is a measure of pool habitat quantity. To learn more about using fish habitat data, see the Habitat Typing background pages.

Downstream Migrant Trapping Data

Downstream migrant traps can yield uniquely valuable information on the productivity and health of watersheds by measuring the number of juvenile fish leaving a stream and providing an opportunity to measure the size of those downstream migrants. Coho and steelhead juveniles must reach at least one year of age, and attain sufficient size, in order to survive the smolt and early ocean life-stages.  Large numbers of coho or steelhead yearlings (1+) indicate a potentially productive salmonid stream, while large numbers of young-of-the-year fish (0+) can indicate habitat conditions too poor to support rearing fish.  Maintaining traps and generating reliable estimates of total out migrant populations is extremely difficult. The California Department of Fish and Game placed and maintained a fyke-net style downstream migrant trap in Caspar Creek in 1999.   Data is presented in two forms, one to illustrate the various types of vertebrates captured at the trap sites, and the other to illustrate the timing and magnitude of out migration for the two age classes of coho and steelhead.

Graves and Burns (1970) operated a cranberry type out migrant trap in the South Fork Caspar Creek in 1964 and 1968 to measure the effect of intervening logging on the emigrating fish community.  Changes in catch may be more the result of changes in trap efficiency than changes in population abundance.  Changes in the relative proportions of fish types captured, however, may reflect changes in fish community structure. 

 For more in-depth information see also Fish Population and Fish Population Information in KRIS.

Large Wood Data from California Department of Fish and Game and Mendocino Redwood Company

The amount of large wood stored in stream channels is an important description of habitat conditions for anadromous salmonids.  Two data sets on large wood in streams of the Big River area and contained in KRIS.  Unfortunately, these data were collected by different protocols and reported using different attributes, so not all comparisons are possible.

The California Department of Fish and Game conducted large wood inventories of Parlin Creek, Hare Creek, and Caspar Creek in 1999.  Data from the later two streams is contained in KRIS Big River.   The CDFG large woody debris and riparian protocol tallies large wood by eight size categories, and attempts to quantify recruitment potential by surveying an area 50 feet outside of the bankfull channel on each bank including conifer or broadleaf trees greater than 1 foot in diameter. The DFG protocol is contained within the Salmonid Habitat Restoration Manual (CDFG 1998). Data from Caspar and Hare Creeks is summarized in the individual reports contained as appendixes to stream reports in the KRIS Bibliography (Hare Creek 1999, Caspar Creek 1999). The Caspar and Hare Creek surveys were done in conjunction with large wood additions. Each stream was divided into a control reach (no wood added) and a treatment reach. The effect of wood on the treatment reach, however, is not described in the inventory reports.   

Mendocino Redwood Company performed large wood surveys of their own design in 2000 and provided data for inclusion in KRIS Big River.  The surveys covered 44 segments from 28 streams across MRC lands in the Big River basin.  The segments measured 20-30 bankfull channel widths in length, and thus ranged from 60-300 meters.  All wood within the bankfull channel was counted and measured if deemed to provide some habitat or morphologic function in the stream channel (i.e. pool formation, scour, debris dam, bank stabilization, or gravel storage).  Wood pieces greater than 12 inches in diameter and 20 feet long were recorded as key pieces if bankfull channel width was less than 20 feet.  In wider stream segments, a larger minimum size was used to classify key pieces.  Debris accumulations (3-10 pieces) and debris jams (>10 pieces) were counted and measured separately. Click here for more detailed information on   MRC large wood survey methods.

For more information on the role of large wood in providing habitat see Large Wood Background page.  The California Department of Fish and Game began stream clearance projects in the late 1950s to try and reverse impacts of sedimentation from logging and to facilitate fish passage. The Mendocino Redwood Company, and its predecessor Louisiana Pacific Corp, have studied large wood removal and calculated the amount of large wood removed in board. See the Stream Clearance Background page to learn more.

Caspar Creek Watershed Studies by USFS Redwood Sciences Lab

The Redwood Sciences Laboratory of the USFS Pacific Southwest Research Station, in cooperation with Jackson Demonstration State Forest, has been investigating watershed processes and effects of timber harvest in the Caspar Creek basin since the commencement of streamflow and sediment monitoring in 1963.  At that time, both the North Fork (473-ha) and South Fork (424-ha) basins of Caspar Creek supported a 90-year-old second-growth redwood and mixed conifer forest.  In summer 1967, a main-haul logging road and main spurs were built in the South Fork with immediate detrimental effects on fish populations (Burns 1968).  The first of three stages of logging began in the South Fork in 1971, during which 59% of the stand volume was selectively cut from 101 ha. In 1972, 69% of the stand volume was selectively cut and tractor yarded from an additional 128 ha. In 1973, 65% of the stand volume was selectively cut from the remaining 176 ha (Rice et al., 1979).

In the North Fork, current era logging occurred in 1985 (67% of an 87-hectare un-gauged tributary was clear-cut and cable), and in the period from 1989-1991 48% of the area was clear-cut and 4% of the streamside protection zone was selectively cut. The timber volume removed from the North Fork was intended to approximate the volume cut from the South Fork in the early 1970's, but clear cutting with cable yarding was used in the North Fork rather than the selective harvest with tractor yarding. Three tributaries in the North Fork were left in an untreated control condition. 

Post-logging measurements continue in the North Fork and South Fork basins to the present.  For the past 38 years, researchers have been studying the nature of hydrologic, erosion, and sedimentation impacts of logging operations.  Turbidity Threshold Sampling Studies have been in operation on both forks since 1995.  Threshold sampling is an automated procedure for measuring turbidity and sampling suspended sediment.

Caspar Creek data on rainfall, runoff, suspended sediment, turbidity, cross-sections and V* from have been included in KRIS Big River to facilitate analysis concerning the effects of timber harvest sediment yields on anadromous fish populations.  Despite the proximity of Caspar Creek to the Big River, cumulative effects from timber harvest could be much more substantial in the Big River watershed basin due to differing geology, micro-climate, and land use.

Research reports on stream channel conditions, sediment yield, and fish populations are available at Caspar Creek Watershed Study. Go to the USFS Redwood Sciences Lab web site for more information. You must be connected to the internet to make this link.

Bulk Gravel Sampling by Hawthorne Timber Company:  Wet-Sieved McNeil Samples

Hawthorne Timber Company (HTC) acquired holdings of Georgia Pacific (GP) Corporation in 1999, so several of the studies cited herein (Ambrose et al., 1996; Ambrose and Hines, 1997; Ambrose and Hines, 1998; Hines, 2000) were conducted under the previous ownership. GP and HTC  have monitored fine sediment at 2 sites in the Big River since 1996 (see Map).  Sediment was collected using a McNeil grab sampler.  Data were provided to the KRIS Big River project in raw form as Excel spreadsheets.  Interpretation of these data was also provided by company biologists in the reports cited above which can be found in the KRIS Bibliography. 

The HTC methods for McNeil sampling follow those recommended by Valentine (1995, in Taylor, Ed. 1996), and the Timber-Fish-Wildlife Ambient Monitoring Program Manual (Schuett-Hames et al. 1994).  Samples were collected with a modified McNeil sampler (modified with a Koski plunger to avoid loss of core material) with a core measuring 15.5 centimeters (cm) in diameter, 13.5 cm in length and capable of holding 2547 cubic centimeters (cc) of material.  Samples were taken from the pool/riffle juncture during the late summer and early fall low flows.  Two riffles were sampled at each station, with four cores taken at each riffle, for a total of eight cores per station.  Individual core samples were averaged, geometric mean and fredle index were then calculated.  To classify the overall particle-size distribution of the sample, based on a geometric progression, eight 30.5 cm frame-diameter sieves were used (63.0, 31.5, 16.0, 8.0, 4.0, 2.0, 1.0, and 0.85mm) ( Shirazi et al. 1981).  

The proposed Big River TMDL threshold threshold for fine sediment included a 6.4mm class size which was not reported by HTC, so their 4 mm class was used instead. This slightly under-represents the true percentage of fine sediment relative to the TMDL threshold because it does not include the cumulative percent of fines between 4 and 6.4mm.  Cores were wet sieved (volumetric method) and correction factors were not applied, causing a slight over-representation of fine sediment as water is increasingly retained with the finer fraction of a bulk sample.  In-stream characteristics noted during collection were stream gradient and stream flow.  As recommended by Valentine (1995, in Taylor, Ed.  1996), measurements were taken along the second medial axis of the three largest rocks collected per individual core.  If the largest particles were greater than 1\3 - 1\4 the diameter of the sampling core, a larger sampler was suggested (ibid).

Matthews and Associates (2001) Preliminary Sediment Budget for Big River

Graham Matthews & Associates (2001) prepared a sediment source analysis and preliminary sediment budget for the Big River watershed. The purpose was to assist the EPA in establishing a Total Maximum Daily Load (TMDL) for sediment in the Big River. The study is primarily based on analysis of aerial photographs and electronic mapping data (GIS), with limited field reconnaissance surveys.  Landslide mapping utilized six sets of aerial photographs: 1936, 1952, 1965, 1978, 1988, and 2000.  Study periods were assumed to extend back to the previous air photo with the earliest period extending back to 1921.  The earliest set, 1936, did not contain coverage for the eastern half of the watershed above the confluence of the North Fork Big River . Surface erosion was calculated for roads and harvest areas following methods developed by Reid (1981) with ridge roads assumed non-delivering. Timber harvests by period were also mapped from aerial photos by Matthews and Assoc. (2001) but harvests prior to 1920 were not discernable. More recent harvest data relied upon information from the California Department of Forestry.

Bulk Gravel Sampling by Graham Matthews Associates:  Dry-Sieved McNeil Samples 

Graham Matthews and Associates (2001) conducted bulk gravel samples (McNeil and Ahnell,1964) at eleven sites in the Big River basin in 2001.  Data from GMA were supplied by the NCRWQCB but information describing particular methods used were not available.  GMA staff confirmed that a modified McNeil sampler (a 12" cylinder placed on the streambed and worked downward as the sample is manually removed) was used at two locations on a single riffle/pool transition representing each site.  Samples were completely dried and weighed producing gravimetric results.  KRIS Big River presents the GMA data similarly to MRC dry-sieve/gravimetric results.  See section below.  GMA did not use a 6.4 mm sieve, but their results for fine sediment <5.6 mm are charted in comparison to a reference value for the <6.4 mm fraction and represent a conservative view of percent fine sediment <6.4 mm.

For more information on how sediment effects fish, and the basis for percent fines thresholds used in KRIS Big River charts, see the Sediment background page. For more on methodology, see Measuring Sediment in Streams.

Bulk Gravel Sampling by Mendocino Redwood Company:  Dry-Sieved McNeil Samples 

Mendocino Redwood Company (MRC) studied bed substrate conditions using bulk gravel samples (McNeil and Ahnell,1964) in five stream segments in the Big River Basin: Daugherty Creek, the East Branch of the North Fork, the Mainstem, the South Fork and Ramon Creek.  These five segments were also subject to gravel permeability sampling, three cross sectional surveys and a thalweg longitudinal profile.  Each stream segment was 20-30 bankfull channel widths in length.  Substrate samples were taken from four randomly selected pool tail-outs in each segment from all pool tail-outs suitable for spawning (i.e., not dominated by bedrock or covered in substrate too large for a fish to make a redd).  For detailed methods, see MRC Sediment Methods.

Bulk gravel samples were taken using a modified McNeil sampler (a 12" cylinder placed on the streambed and worked downward as the sample is manually removed).  MRC dried their gravel samples before passing through 7 different size-class screens (45, 22.4, 11.2, 5.6, 4, 2, 0.85) and then weighed each fraction to produce gravimetric results.  This dry-sieve/gravimetric method is the same as used by Graham Matthews and Associates, but produces different results than the wet-sieve/volumetric method used by Hawthorne Timber Company and originally proposed by McNeil and Ahnell (1964).  While the dry-sieve/gravimetric method produces results not biased by the retention of different proportions of water by size class ( water is increasingly retained with finer fractions of a bulk sample),  results from the method are not directly compatible with TMDL thresholds and other synthesis of wet-sieve based studies. The U.S. EPA and California State Water Resources Control Board do not necessarily endorse the conversions used in KRIS described below (see Sediment Background page).

KRIS Big River uses reference values for percent fine sediment from dry-sieve sampling that utilize a conservative adjustment to be more compatible with wet-sieve/volumetric results.  The Big River TMDL threshold of no more than 14% fine sediment less than  0.85 mm comes from a synthesis of studies using wet- sieve bulk gravel sampling.  According to Shirazi and Seim (1979), the fraction of fines less than 0.85 mm from wet/volumetric methods can be adjusted by a factor of 0.739 to reflect actual gravimetric results.  Applied to 14%, this correction yields a dry- sieve reference for fine sediment less than 0.85 mm of 10.3%.  McHenry et al. (1992) found an even higher conversion factor for wet to dry sieve comparison. For the fraction of fines less than 6.4 mm from wet/volumetric methods, Shirazi and Seim report a correction factor of 0.866.  Applied to the TMDL threshold of 30%, this correction yields a dry- sieve reference for fine sediment less than 6.4 mm of 26%. MRC did not use a 6.4 mm sieve, but their results for fine sediment <5.6 mm are charted in comparison to this reference value and represent a conservative view of percent fine sediment <6.4 mm.

For more information on how sediment effects fish, and the basis for percent fines thresholds used in KRIS Big River charts, see the Sediment Background page. For more on methodology, be sure to see Measuring Sediment in Streams.

Gravel Permeability Sampling by Mendocino Redwood Company

Gravel permeability in the Big River was measured by Mendocino Redwood Company following the methods of Barnard and McBain (1994). A stand-pipe was driven into the streambed to a depth of 25 centimeters and vacuumed by an electric pump. The units of measurement for gravel permeability are cm/hr and describe interstitial flow rate.  At each measurement location, repetitive measurements were taken until the permeability readings ceased to increase.  The results reported here may vary slightly from those published by MRC due to the different treatment of summary statistics.  Five segments were studied in the Big River basin: the mainstem, the East branch of the North Fork, the South Fork,  Ramon Creek and Daugherty Creek.  Each of the five segments studied had between four and ten pool tails or “tail-outs” with between two and eight grid locations in each tail-out.  A grid location consisted of an evenly spaced 12-point grid with between four and ten separate randomly selected measurement points.  This sampling design led to six hundred and thirty seven permeability measurements from 125 grid locations in 34 tail-outs in the year 2000.  Individual measurements from each grid location were averaged and used as independent data points in this analysis.  The summary statistics presented in KRIS were derived from averaging the individual measurements from each grid point, then averaging the data collected from each grid, then averaging each tail-out, and finally averaging all the tail-outs to represent the permeability for a stream segment.

The natural log of permeability derived by Tagart (1976) and McCuddin (1977) (Stillwater Sciences, 2000) equates the permeability data to fry survival (Survival = -0.82530 + 0.14882 * ln permeability).  In a few cases, the survival index was a negative number.  In these cases, the index was reported as zero.  The survival relationship is an index of spawning gravel quality and is currently one of the few approaches that quantitatively links a biological relationship to permeability data.

For more information on how sediment effects fish, and the basis for percent fines thresholds used in KRIS Big River charts, see the Sediment Background page. For more on methodology, be sure to see Measuring Sediment in Streams.

Knopp North Coast Regional Sediment Study

Excessive fine sediment fills pools and causes general instability of stream gravels and channel form.  Loss of pool depth and reduction of channel complexity lowers the carrying capacity of streams for aquatic biota.  Knopp (1993) sought to determine which physical elements of in-stream habitat were affected by human activity and to measure the range of values for those elements in disturbed and un-disturbed streams at 60 northwestern California sites.  Knopp found that sediment in streams is positively correlated to watershed disturbance.  Knopp also found that V* (V-star) is suitable for assessing the amount of fine sediment in pools. V-star is the proportion of a pool's residual volume filled by fine sediment (Hilton and Lisle, 1993) and V* values from Mendocino streams, including eight streams in the KRIS Big River planning area are presented in KRIS Big River.  Data from Knopp's testing of other indices (large wood, substrate particle counts, pool statistics and pool frequency) are available in the KRIS source table, Knopp.dbf.  See the Sediment Background page "Sediment Information in KRIS" to learn more, including about the 0.27 V* threshold shown on the charts. 

Jackson State Forest Sediment Monitoring

The Jackson State Demonstration Forest (JDSF) comprises large portions of the Little North Fork, and North Fork sub-basins of the Big River, as well as Caspar Creek and other coastal watersheds in the area.  In 1993, JDSF personnel repeated measurements of stream gravels originally performed by Burns in the 1960's (Burns 1970).  

Burns (1970, 1971, 1972) evaluated logging impacts on salmonid habitat and populations in northern California streams during the late 1960's.  He investigated watersheds that had been logged including Bummer Lake Creek, South Fork Yager Creek, South Fork Caspar Creek, and Little North Fork Noyo River.  His results showed that extensive use of bulldozers on steep slopes or in stream channels can cause excessive erosion which may be detrimental to salmonids. Also he suggested that warming of waters through canopy opening may be a problem.

The JDSF study was never published, so the methods are not clear, but are likely to be the same as those of Valentine and Jameson (1994).  To summarize, "bulk samples were collected from the thalweg where the water's surface velocity noticeably increased (breaks) as it passed from the tail of a pool into a riffle.  A single sample was collected from five pool/riffle breaks in each study reach. Where this sampling location was clearly not suitable for spawning as determined from the substrate's surface conditions, a sample was not taken... and the sample was moved to the next upstream pool / riffle break.  Samples were collected using two McNeil samplers (McNeil and Ahnell 1960) with core dimensions of 14.4 cm ID x 13.4 cm and 15.2 ID cm x 15.2 cm, resulting in nominal sample volumes of 2183 and 2780 cm3 respectively.  Samples were wet-sieved (Armour et al. 1983:31) in the field following Valentine (1993), including a 10 minute settling time for fines (<0.85 mm). Substrate metrics calculated from the samples include percent of the samples passing the 0.85 mm sieve ("fines"), the percent passing the 3.3 mm sieve, the geometric mean, and the Fredle index (Armor et al. 1983).  The value passing the 3.3 mm sieve was calculated mathematically using a logarithmic conversion of particle sizes because natural sediments frequently exhibit lognormal distributions (Platts et al. 1979)."

Timber Harvest Information in KRIS Big River

Timber harvest data in KRIS comes from the California Department of Forestry (CDF). Timber harvest information is displayed in the KRIS database as images of maps but full spatial data is available in the KRIS Big River Map project. Data can be accessed in Arc View or in Arc Explorer, which is free-ware included on the KRIS Map CD. Further analysis of timber harvest is taking place related to the Big River TMDL, which was not available as this version of KRIS Big River went to press. Read more about the Timber Harvest  maps in the KRIS Big River project.

Big River Flow and Regional Rainfall Data

Flow data in KRIS Big River comes from U.S. Geologic Survey (USGS) records for the South Fork Big River near Comptche gage (#11468070) and Graham Matthews and Associates as part of the data acquired for the Sediment Source Analysis and Preliminary Sediment Budget for the Big River Watershed (Matthews, 2001).  USGS records of average daily flow are in cubic feet per second and extend from October 1, 1960 to September 30,1971.  Matthews developed synthetic streamflow data to extend the USGS record using USGS Noyo River flow data.  Additionally, Matthews collected continuous streamflow data at four stations in the Big River between November 2000 and April 2001.  

Rainfall in KRIS Big River is available for three gauging sites, including: two gages in Mendocino, CA and one at Russian Gulch State Park.  Data origin is the National Weather Service NCDC database and James Goodridge, former state climatologist and now consultant to the California Department of Water Resources. Rainfall data records extend from 1978 to 2000 and contain some data gaps at individual sites.  See the Stream Flow  background page for more information.

Historical Photos Courtesy of the Held-Poage Library and Mendocino Historical Society

Historical photos in the KRIS Big River project were scanned from the collection at the Held-Poage Museum and Library in Ukiah. This treasure trove of historical images allows us to see channel conditions and land and gain valuable insight into an era before scientific data was collected. The images are for viewing in this project only and further use of these photos requires the express and written permission of the Mendocino Historical Society. Photo negatives were collected by Robert Lee who developed the prints on display at Held-Poage. Photos date back as early as the 1860s. Notes are from Mr. Lee and numbers in his reference system are entered in the captions for the photos.

KRIS Map Project Images

Images of maps in the KRIS Big River database are from associated Arc View and Arc Explorer projects. They are made available for convenience but full spatial data is available. See KRIS Big River Map background page or review projects on the KRIS Map CD. Full Arc View projects often use the Spatial Analyst but are also viewable without that extension with the grid data converted to an image.

Coastal Tributaries Fisheries Resources

California Department of Fish and Game file memos supply quite a bit of information about the smaller coastal tributaries that lie between the Noyo and Big Rivers. While Hare and Caspar Creeks have received much study and attention, streams such as Mitchell (2, 3) and Doyle Creeks and Russian Gulch (2, 3, 4) have received less attention. The links under the stream names connect you with CDFG file reports.

Aerial and Ground Photos Contributed by Nicholas Wilson

Nicholas Wilson contributed aerial and ground photos of timber harvests and watershed conditions from both the 1970s and the 1990s to the KRIS Big River project. Nick is a photographer who lives in Little River, California. All of his photos are copyrighted and can only be used with his permission. Nicholas Wilson,707-937-0137, PO Box 943, Mendocino CA. Nick's original captions and notes on photo processing can be viewed as File Info inside Photoshop if photos from KRIS are opened in that program.

Bancroft Library Digital Photographs

"The Bancroft Library is the primary special collections library at the University of California, Berkeley. One of the largest and most heavily used libraries of manuscripts, rare books, and unique materials in the United States, Bancroft supports major research and instructional activities and plays a leading role in the development of the University's research collections" The Bancroft Library allowed use of Big River photographs from the Watkins Collection, which it has posted on-line.

Aerial and Ground Photos Contributed by Coast Geographics

Rixanne Wehren of Albion shared aerial and ground photos of the Big River watershed that appear in KRIS Big River. Rixanne does work related to watersheds and restoration, including work in Arc View, through Coast Geographics, P.O. Box 340, Albion, CA 95410, (707) 937-2709.

Vegetation and Timber Types of Calwater Watersheds

The vegetation and timber types used in the KRIS Noyo project was derived from Landsat multi-spectral images taken in 1994. The U.S. Forest Service Pacific Southwest Region Remote Sensing Lab, in cooperation with the California Department of Forestry, analyzed the Landsat images to formulate a California-wide electronic map layer of vegetation as part of the Northwest Forest Plan (Warbington et al., 1998). See Vegetation Types Background page for more information)

Stand conditions are accurately represented at the one hectare scale by the USFS vegetation data. Data is quarried for tree size or community type in KRIS Noyo Maps. This allows quantitative assessment of vegetation types for seral stage based on tree size for geographic areas such as Calwater planning watersheds in KRIS DB.

For use in KRIS, vegetation and timber types were simplified into ten classifications. Vegetation classifications are:

Very Large Trees = 40" in diameter or greater
Large Trees = 30-39.9" in diameter
Medium/Large Trees = 20-29.9" in diameter
Small/Medium Trees = 12-19.9" in diameter
Small Trees = 5-11.9" in diameter
Saplings = 1-4.9" in diameter
Non-Forest = Non-tree species such as shrubs, grasses or bare soil

This simpler classification provides an easy to understand index of watershed disturbance for use in coastal watersheds. Large components of early seral stage conditions (Saplings, Non-Forest) are often associated with recent logging disturbance. The vegetation patterns in interior basins, such as the Eel River and Klamath River watersheds, are much more complicated than in coastal ecosystems and more difficult to use to analyze changes in vegetation brought about by watershed management activities. The KRIS vegetation classification scheme can also be used for a quick analysis of riparian conditions. Ninety meter (297 ft.) zones of riparian influence are assigned to the 1:24000 stream layer in Arc View and only the vegetation within this zone is displayed and analyzed.


Note about documents in KRIS