Warning This Web page has been archived on the Web.

Archived Content

Information identified as archived on the Web is for reference, research or recordkeeping purposes. It has not been altered or updated after the date of archiving. Web pages that are archived on the Web are not subject to the Government of Canada Web Standards, as per the Policy on Communications and Federal Identity.

Skip booklet index and go to page content

Revised Recovery Strategy for the Nooksack Dace (Rhinichthys cataractae) in Canada [Proposed]

3. Critical Habitat

3. Critical Habitat


3.1.      Definition

Critical habitat was defined using in-stream habitat characteristics at the scale of the reach, a natural unit of stream habitat that ranges from hundreds to thousands of metres in length (Frissell et al. 1986). There are three reasons for adopting this scale. First, the reach scale corresponds to the distribution of subpopulations of both species within watersheds and usually contains all habitat types used during the life history cycle (Pearson 2004a). Second, the ‘channel units’ of critical habitat (riffles and pools) are dynamic and frequently move during flood events in these streams. Effective protection and management of critical habitat in these circumstances must allow for normal channel processes and must, therefore, occur at a spatial scale larger than the channel unit. The reach scale is the next largest in accepted stream habitat classifications (Frissell et al. 1986; Imhof et al. 1996) and by definition represents relatively homogenous segments of stream demarcated by distinct geomorphic or land use transitions. Third, the reach scale corresponds most closely to that of land ownership in these watersheds and, consequently, to most recovery actions.

Critical habitat includes riparian reserve areas. Reserve widths are assessed using a GIS based methodology adapted directly from and consistent with that of the British Columbia Riparian Area Regulation (RAR, Reg. 837 under the Fish Protection Act (S.B.C. 1997, c. 21) Anonymous 2005).  The width of existing riparian vegetation and areas where riparian reserve width is restricted by permanent structures (roads, buildings, yards etc.) are also mapped.  

3.2.      Identification of Critical Habitat

Critical habitat includes all habitats within occupied watersheds that the Recovery Team considers of high quality or potentially high quality for Nooksack dace, and constitutes the habitat that the Recovery Team deems necessary for species persistence, and to achieve recovery objectives. Critical Habitat for Nooksack dace consists of reaches in their native creeks and that consist of (or are known to have previously consisted of) more than 10% riffle by length. It includes all aquatic habitats and riparian reserve strips of native vegetation on both banks for the entire length of the reach. Reserve strips are continuous and extend laterally from the top of bank to a width equal to the widest zone of sensitivity (ZOS) calculated for each of five riparian features, functions and conditions: large woody debris supply for fish habitat and maintenance of channel morphology, localized bank stability, channel movement, shade, and insect and debris fall. The ZOS values are calculated using methods consistent with those used under the British Columbia Riparian Areas Regulation (Reg. 837) under the Fish Protection Act (S.B.C. 1997, c. 21).

The combined length of critical habitat for Nooksack dace is 33.1 km (of 93.9 km of surveyed stream channel). Maps showing the extent of critical habitat for the watersheds known to contain Nooksack dace are provided in Appendix 2.

3.3. Rationale

Riffle Habitat

Available information overwhelmingly indicates that Nooksack dace require riffle habitats and that reaches with a high percentage of riffle habitats support most of the population. Nooksack dace typically occur in riffles with loose gravel and cobble substrates where water velocity exceeds 0.25 m.s-1. They spawn near the upstream end of riffles (McPhail 1997) between late April and early July (Pearson 2004a) and forage nocturnally for riffle dwelling insects (McPhail 1997). Logistic regression relating Nooksack dace presence to habitat type (riffle, shallow pool etc.), cover availability and riparian land use showed that reach occupancy was most strongly predicted by the amount of riffle habitat present, and that riffles isolated by long stretches of deep pool are seldom inhabited (Pearson 2004a). The proposed threshold of 10% riffle by length is intended to exclude reaches with very small amounts of riffle habitat that contribute minimally to Nooksack dace production and population size.

A number of reaches containing less than 10% riffle by length when surveyed are included in critical habitat (Table 4) because of evidence that they previously contained more riffle habitat and supported Nooksack dace populations. Most of these reaches are known to have been channelized and dredged or were temporarily impounded by beaver at the time of survey. All currently contain Nooksack dace except four reaches in Fishtrap Creek. These are known to have contained abundant riffle and Nooksack dace prior to dredging (J.D. McPhail pers. comm.). The remaining (non-critical habitat) reaches in all watersheds contain a total of 490 m2 of riffle habitat, or 1.9% of the total riffle habitat present.

Shallow Pool Habitat

Young-of-the-year Nooksack dace inhabit shallow (10-20 cm) pools adjacent to riffles where they swim above sand, mud, or leaf litter substrates and feed upon chironomid pupae and ostracods (McPhail 1997). Insofar as these habitats are exclusively used for larval rearing before juveniles move in to riffle habitat, the loss of these habitats would likely cause population declines.

Riparian Habitat

Riparian vegetation is included in critical habitat to the extent necessary to protect the integrity of in-stream critical habitat. Loss of riparian vegetation will result in bank erosion, siltation, water temperature elevation, and nutrient inputs that will directly degrade instream critical habitat.  Required widths will vary among sites and are defined in reach scale assessments.  Reserves must be sufficient to control sediment entry to the stream from overland flow, to prevent excessive bank erosion and to buffer stream temperatures. Reserve areas will also remove significant amounts of nitrate and phosphorous from groundwater, although their efficiency depends strongly on hydrogeologic conditions (Martin et al. 1999; Puckett 2004; Wigington et al. 2003). The effectiveness of a riparian reserve in preventing materials (sediment, nutrients, toxins, etc.) from entering a stream depends upon on its continuity in addition to its width, particularly when it is narrow (Weller et al. 1998). Consequently, riparian reserves in critical habitat reaches should be continuous. In open landscapes, such as agricultural fields, vegetation from reserve areas will collect windblown insects (Whitaker et al. 2000). Such insects,falling from riparian vegetation into the water constitute an important food source in headwater streams (Allan et al. 2003; Schlosser 1991).  More than 30 m of riparian vegetation may be required for full mitigation of warming (Brown & Krygier 1970; Castelle et al. 1994; Lynch et al. 1984), and siltation (Davies & Nelson 1994; Kiffney et al. 2003; Moring 1982), and for long-term maintenance of channel morphology (Murphy et al. 1986; Murphy & Koski 1989). At least 10 m are required to maintain levels of terrestrial food inputs similar to those of forested landscapes (Culp & Davies 1983). Reserves as narrow as 5 m provide significant protection from bank erosion and sediment deposition from overland flow (Lee et al. 2003; McKergow et al. 2003).

Failure to maintain an adequate riparian reserve as part of critical habitat is likely to cause population-level impacts. In habitats lacking sufficient flow or groundwater, absence of shade may increase water temperatures to harmful levels, especially under climate warming scenarios. Increased erosion due to poor bank stability will cause direct sediment deposition in riffles, impairing spawning and incubation, reducing food availability, and eliminating the interstitial spaces in coarse substrate that Nooksack dace and their prey occupy. Nutrient loading will be higher in reaches without adequate riparian vegetation (Dhondt et al. 2002; Lee et al. 2003; Martin et al. 1999) and is likely to contribute to hypoxia through eutrophication. Increased solar radiation in nutrient rich reaches lacking adequate riparian shading (Kiffney et al. 2003) will also contribute to eutrophication and hypoxia. 

Width of riparian reserves required to protect key habitat attributes for Nooksack dace have not been quantified. R. cataractae is certainly less dependant upon deep pool habitats than salmonids are, suggesting somewhat lesser requirements for large woody debris. They also favour benthic over drifting invertebrates (Scott & Crossman 1973) suggesting they are less dependant on insects of terrestrial origin. R. cataractae appear tolerant of slightly higher water temperatures than salmonids (Wehrly et al. 2003), suggesting a reduced need for shading, but this may not be true under future climate warming scenarios. However, Nooksack dace are likely to be equally or more vulnerable than salmonids to habitat degradation caused by sedimentation, loss of scope for natural channel movement, and invasive plant overgrowth of riffles fuelled by nutrient loading and riparian loss. Benthic insectivores and fluvial specialists, like Nooksack dace, are among the most sensitive fish species to loss of wooded riparian areas (Stauffer et al. 2000), probably  due to the impacts of siltation and alterations to macroinvertebrate community structure (Allan 2004; Kiffney et al. 2003). Overall, there is little reason to believe that Nooksack dace require narrower buffers than salmonids.

BC MOE and DFO have developed and implemented a methodology for determining riparian reserve widths required to protect fish habitat in streams that they deem to be minimally sufficient in maintaining riparian function to protect fish habitat. The Riparian Area Regulation (RAR) was developed under the Fish Protection Act to protect “salmonids, game fish, and regionally significant fish” from the impacts of land development. In the absence of definitive data for a SARA listed species, this seem to be a reasonable standard to apply in the identification of critical habitat, as it represents a benchmark and standard methodology to which both federal and provincial agencies responsible for management of species at risk have already agreed, and it forms the basis of the methodology employed (see below).  The width of riparian buffers sufficient to protect fish habitat is a scientific discipline in itself, and it is neither practical nor within the mandate of the Recovery Team to develop an independent assessment methodology and regulatory framework.

Finally, it should be noted that unidirectional transport of sediment in flowing waters means that riparian reserve strips upstream of critical habitat reaches are important in minimizing sedimentation and other impacts within critical habitat. For this reason stewardship programs should promote the establishment of continuous riparian reserve strips of native vegetation throughout the watershed, not just along critical habitat reaches. 

Table 4: Reaches included in critical habitat for Nooksack dace that contained less than 10% riffle by length at the time of survey (1999).

WatershedReachLengthRiffle LengthRiffle Area% Riffle by LengthDace PresentCondition
BertrandBTD5652401126.1YChannelized and dredged
 BTD744929586.5YPartially impounded by beaver
 BTD81139441763.9YPartially impounded by beaver
FishtrapFTP119841704598.6NDredged 1990-1991
 FTP21239721445.8YDredged 1990-1991
 FTP396215331.6NDredged 1990-1991
 FTP6926661987.1NDredged 1990-1991
PepinPEP12635131.9YChannelized and dredged

3.4. Activities Likely to Result in Destruction of Critical Habitat

Many of the threats that face Nooksack dace are habitat-related, and this is of particular concern. Threats to critical habitat for Nooksack dace are identified, and the reader is also referred to the Threats section for a more thorough discussion of threats identified below. It is also important to note that there are many gaps in our understanding of critical habitat features and their threats, and that this will be a focus for research in one or more action plans.

Excessive water withdrawalWater extraction (surface or ground) during dry periods reduces flows, which may contribute to hypoxia and drying of riffles needed for spawning.
Excessive sediment releasesSediment deposition in spawning substrate and inhibition of the flow of oxygen-rich water to eggs and larvae during incubation.
Drainage projectsDredging, dyking, and channelization works directly destroy habitat, cause sediment deposition in riffles, and reduce base flow,
ImpoundmentPonding caused by either human or beaver activities eliminated riffle habitat.
Urban storm drainageStorm drain systems that discharge directly to creeks are major sources of toxic contamination and sediment. They also reduce baseflow by inhibiting water infiltration to aquifers.
Riparian vegetation removalRiparian vegetation removal exposes a stream to increased erosion and sediment deposition, elevated water temperatures, reduced supplies of terrestrially derived food, and increased nutrient loading
Livestock access to creeksLivestock damage habitat by trampling or causing erosion that clogs riffles with sediment. Access also contributes to nutrient loading.


ActivityBertrandCreekPepin BrookFishtrap CreekBrunetteRiver
Excessive water withdrawal++++++?
Excessive sediment releases++++++?
Drainage projects++++++?
Urban storm drainage+++-++++++
Riparian vegetation removal++++++?
Livestock access to creeks+++?


+++major concern+minor concern
++moderate concern-/?not a concern/unknown


3.5.  Schedule of Studies to Identify Critical Habitat

Information exists to assist in the definition of critical habitat for Nooksack dace throughout its presently known range. Further surveys are required to identify other potential populations and characterize their critical habitats, as summarized below:

Population IdentificationThe Coquitlam and Alouette Rivers are suspected of containing Nooksack dace based on a preliminary genetic and morphometric study of their R. cataractae populations (J.D. McPhail, UBC, unpubl. data). Additional samples are required for confirmation.2005-2006Underway
Critical Habitat SurveysHabitat in the BrunetteRiverhas not been surveyed as its populations were unknown prior to 2004.  Surveys will also be required in the Coquitlam and Alouette Rivers if the presence of Nooksack dace is confirmed there.2006-2007Planned

3.6. Knowledge Gaps in Nooksack Dace Biology

Additional studies should be conducted to address the following data needs related to specific threats to Nooksack dace. This information will contribute to the protection of Nooksack dace and their critical habitats.


Impacts of Riffle DryingThe fate of dace in reaches that dewater during late summer is uncertain. Sampling during this period will resolve whether fish leave the reach, move into pools, burrow into substrate, or die.2004-2005Underway
Impacts of Sediment Deposition in RifflesThe extent to which sediment deposited in riffles affects their ability to support healthy dace populations is uncertain and needs to be quantified.2007-2008Need Identified