>Fish Community Diversity

Fish Community Diversity

This indicator measures the diversity in fish species.
Biotic Condition

QT: Indicators

What is it?

The abundance of individual fish and the diversity of fish species present can tell a lot about the conditions in waterways and watersheds. Two metrics of fish community condition were used, based on indicators used by the South East Queensland Ecosystem Health Monitoring Program (Australia). These metrics are:

  • Percentage of native species expected (PONSE). This is a measure of observed number of fish species (species richness) compared to expected number of species based on expert knowledge and observations in other regions of the river. The primary source for expected species was the Sierra Nevada Ecosystem Project (SNEP), with adjustments based on survey effectiveness (see below under “Data manipulation”).
  • Proportion native species (PNS). This is the percentage of native species of total fish caught or observed (not species number). This metric assesses what proportion of the community is composed of native species. Native/exotic identity was determined using information from the SNEP report.

In addition to these fish community composition measures, local Chinook salmon population assessment was used for the Lower Feather and Lower Yuba regions. Population is scaled from 0 to 100 by comparing recent six-year geometric mean of abundance to DFG targets for population restoration (Table 1).

Why is it important?

Fish are a common and familiar component of freshwater environments, and fish communities reflect a range of natural and human-induced disturbances through changes in abundance and species composition. Ecological assessments based on fish community structure have the advantage over more traditional physical and chemical indices (e.g. conductivity, turbidity, nutrients) in that fish provide an integrated measure of stream condition due to the mobility, relatively long-life, and high trophic level of the animals involved (SEQ EHMP methods). Low native species presence can be an indicator of high disturbance levels, which disrupt natural community balance and exclude stress-intolerant species and/or non-generalists. Presence of exotic species is also a good indicator of poor ecological health (Meador et al. 2003). Many invasive species are highly competitive generalists, and can exclude local species. In addition, exotics may be able to establish due to altered habitat processes (i.e. higher water temperatures, changes in mean water level) or through direct human introduction (i.e. stocking, discard of aquarium fish).

Native salmonid species are of great ecological, economic, and cultural importance to local communities. They also serve as strong indicators of habitat quality and integrity in river systems, particularly with regard to water temperature, sediment load, and barriers to passage. The Central Valley spring-run Chinook salmon in particular is listed as a threatened species under the ESA, giving them a high priority for restoration. The main threats to the remaining populations are loss and degradation of habitat. In particular, rising water temperature combined with loss of upstream spawning and rearing habitats blocked by dams has diminished available juvenile summer habitat greatly. Within the Feather River Watershed, only two populations persist. One, in the Feather River itself, is completely dependent on the Feather River FIsh Hatchery to maintain itself. The other, in the Yuba River, is of unknown status.

Target or Desired Condition

Ideally, native fish communities will be fully intact and contain no invasive or introduced species. A PONSE and PNS of 100 indicate that every expected species was found in the area, and no exotic species were caught.

For Chinook populations, the target condition is taken from DFG’s Central Valley targets (DFG 1988). Because the targets are given for the Sacramento Basin as a whole, the five-year geometric mean proportion of total population for the subwatersheds was used to determine a specific restoration target for the Feather and Yuba populations. The target for population for each of the Feather and Yuba was based on the proportion of the total observed returning salmon to the Sacramento River that returned to each of these rivers.

What can influence or stress condition?

Primary stressors for native fish communities are habitat degradation, high fine-sediment load, increasing maximum water temperatures, and introduced species. Salmonids are negatively impacted by increased maximum water temperatures, sediment loads, habitat loss, barriers to passage, and predation by black bass and striped bass on their young.

Data sources

DWR fish community survey data on the Feather River (1997-2005, 6 locations)

FRCRM 2001, 2003 electrofishing data (multiple creeks)

NID, FERC survey data on the Yuba and Bear River (2008-2009)

Friends of Deer Creek fish community survey data (2007-2008)

Vaki Riverwatcher 2008-2010

DFG Annual Reports Chinook Salmon Spawner Stocks in California’s Central Valley (1953-2009)

Data transformation and analysis

Unidentified species

Unidentified species were not relevant to PONSE calculation, and were only included in PNS if the native/exotic status was well-defined (i.e. “unidentified sculpin” or “unidentified bass,” given that only native sculpins and exotic bass are found in the species lists).


Survey data for each site were combined annually, so that each site had only a single data point for each year. Fish abundances from individual surveys were added together to form one “total survey.” Subwatershed trends were then determined via a Regional-Kendall analysis across all sites within the subwatershed. Current state was determined by averaging PONSE and PNS scores for the most recent year at each site within the subwatershed. Most datasets included only a few samples per year in a few months, so monthly analysis was impossible.

Expected native species

Expected native species were determined through a combination of factors. First, a list of species native to the Sacramento River Basin was taken from the SNEP report (SNEP, 1996). Then the subwatersheds were divided into low and high elevation sites, and the native fish were divided accordingly depending on habitat preferences. Elevation decisions were made by estimating whether each subwatershed was primarily lowland/foothills or in the mountainous region of the watershed. Finally, the lists were reduced to only fish that had been caught in at least one of the relevant subwatersheds. This limited the number of expected species to only those known to be found by the electro-fishing surveys employed. Expected species for the Lower Feather were slightly expanded, because the surveys conducted there were more thorough than in other regions. Final expected species lists are given in Table 10.

Salmon abundance

The abundances for the most recent years (2006-2009) are considered preliminary. Confidence in current state is accordingly slightly lower.

What did we find out/How are we doing?

Overall scores for all subwatersheds based on fish community composition and abundance of salmon in the lower watershed are given in Table 2.

Fish Community Composition

Survey information was limited, with only a few surveys and years per subwatershed. Most surveys were performed during only a few months per year, so annual aggregation and trend analysis was most appropriate. Although not all surveys were conducted identically, with combinations of electrofishing, snorkeling, and passive monitoring, results could be compared once converted into PONSE and PNS values. No data were available in the Lower Bear subwatershed. Score calculation was straight-forward, with 100 indicating all expected native species detected and no exotic species caught (with 0 indicating the reverse). Current status for each subwatershed was calculated by averaging across all sites for their most recent years. The PONSE and PNS are given in Tables 3 and 4, respectively. The majority of native species caught were Chinook salmon (in the Lower Feather and Lower Yuba) and Rainbow trout (in all subwatersheds).

Chinook Salmon Population

Chinook populations were analyzed in the Lower Feather and Lower Yuba subwatersheds. Although salmon spawn in the Lower Bear River, no regular monitoring of spawning is conducted there. Annual population sizes for the fall- and spring-run Chinook in both rivers are shown in Figures 2a and b. Current condition was calculated as described above, and the scores are given in Table 7. Note that the Feather River fall-run is composed of about 20% hatchery fish, whereas the spring-run is entirely maintained by hatchery fish. Confidence in these scores is slightly lower, because the data for the most recent years (2006-2009) are considered preliminary. However, these numbers are usually only adjusted slightly so all conclusions should be valid.

Trends were calculated using a Mann-Kendall analysis, and results are given in Table 8. Overall conditions are shown in Table 1, with salmonid scores averaged with the Lower Feather and Lower Yuba fish community scores. Note that the only positive trend was for the spring-run of Lower Feather Chinook. This is despite very low numbers in 2008-2009. The spring-run is an endangered population. Both Feather River populations had a spike in abundance around 2001-2003, and have diminished since. It is unclear whether this is natural variation or due to anthropogenic factors.

Temporal and spatial resolution

There is an uneven distribution of fish monitoring sites across the watershed and a tendency for these sites to be associated with FERC relicensing processes and thus rivers with regulated flows and large dams and reservoirs. The highly-regulated South Yuba had a disproportionately large number of sites relative to other subwatersheds. Because of the association of fish community monitoring with FERC re-licensing, most data have been collected recently. Returning salmon have been well monitored in this watershed for many years, though the methods for assessing population health have changed over the years. Currently there is extensive monitoring of returning adults in the lower watershed, but not as extensive monitoring of juvenile rearing, growth, and return to the ocean.

How sure are we about our findings (Things to keep in mind)

For both fish community and salmon population assessments assumptions were made about targets (species richness and abundance). Modifications of these targets because of habitat variation (fish community metrics) and new estimates of target populations (salmon) would change the corresponding condition scores.

There were sufficient data to be confident about the fish community findings and the spawning salmon abundances to be confident in the trends measured. Longer term monitoring will be needed to estimate trends in fish communities with greater confidence.