Brook Trout Restoration

Over the past 100 years, brook trout, Salvelinus fontinalis, (Figure 1) have experienced extensive declines throughout their native range. Declines have been especially precipitous in the central and southern Appalachian Mountains of the mid-Atlantic and southeastern US.

Numerous causes have been ascribed to brook trout declines, ranging from overharvest (Figure 2), sedimentation, acid precipitation, exotic species introductions, and climate change. In many areas of the central Appalachians, acid precipitation alone has impacted 60–80% of brook trout spawning habitat (Petty and Thorne 2005).

Historically, brook trout conservation was a low priority for resource managers due to the perception of brook trout as a low value fishery relative to introduced salmonids. For example, following widespread forest clear-cutting at the turn of the 19th century in the southern Appalachians, brook trout streams were decimated by overfishing, and significant warming occurred due to a lack of canopy cover (King 1942). Because of their higher tolerance to warmer temperatures, rainbow and brown trout were introduced to replace brook trout as a recreational and sustenance fishery. Many brook trout populations continue to be impacted by historic land cover alterations and competition with non-native trout.

However, as interest in native species conservation has increased, substantial resources are being directed towards brook trout restoration. In recognition of the need to address range-wide threats to brook trout, the Eastern Brook Trout Joint Venture (EBTJV) was formed in 2005. The goals of the EBTJV are to implement coordinated strategies to improve aquatic habitat, raise public awareness, and prioritize the use of federal, state and local funds for brook trout conservation (Figure 3).

An important component of the range-wide brook trout conservation strategy is the implementation of specific actions designed to restore brook trout populations and habitats. Such actions include: re-introduction of brook trout into streams where they have been locally extirpated, implementation of catch-and-release fishing regulations, stream channel reconfiguration and structural habitat enhancement, riparian zone re-vegetation and fencing, acid remediation through limestone sand application, removal of dispersal barriers, and removal of exotic species.

Despite widespread implementation, it is largely unknown how restoration activities affect brook trout population processes occurring at the watershed scale. Furthermore, we do not know the extent to which restoration effectiveness will persist under a changing climate in which many threats to brook trout survival are magnified. The objectives of this article are to: 1) describe the overall process of ecological restoration as it relates to recovery of brook trout and Appalachian coldwater streams; 2) summarize the types and effectiveness of restoration approaches that are currently undertaken; and 3) discuss needs and opportunities for improved brook trout restoration effectiveness as we confront a changing climate.

Ecological restoration should be viewed as a form of adaptive management where specific restorative actions represent experimental treatments (Roni 2005, Palmer et al. 2005). In this way, restoration actions are prescribed trials that can be assessed quantitatively relative to a priori expectations and objectives for success. Specific restoration approaches can then be judged in terms of their benefits and shortcomings under different environmental settings. For our purposes here, we are focused on establishing an adaptive process capable of producing more effective brook trout restoration programs over time.

Key elements of the process include:

Brook trout restoration approaches, and species restoration in general, can be divided into two broad types of actions: 1— actions designed to affect the targeted population directly, and 2 — actions designed to affect populations indirectly through amelioration or enhancement of the structural, hydrologic, chemical, and "biological" (e.g., competitors or parasites) habitat.

Species re-introduction. Increasingly, conservationists are attempting to re-establish brook trout populations into streams where they have been previously extirpated. In fact, many conservation and state agencies include re-introduction efforts in brook trout restoration plans (e.g. WVBTCG 2006; ODNR 2007). The purpose of these efforts is to restore brook trout populations that are able to sustain themselves through in-situ survival, growth, and reproduction. This restoration technique requires that current physical and chemical conditions are suitable for survival and that the underlying factors originally resulting in extirpation have been identified and adequately controlled. For example, Hudy et al. (2000) were able to successfully re-introduce brook trout into a stream following in-stream acid remediation. An important pitfall of species re-introduction, however, is that it has the potential to negatively affect regional population genetics. This occurs in two ways. First, the individuals that are introduced into a particular stream may have a genetic make-up that is substantially different from the native, intact populations in the region, which in turn can result in outbreeding depression and/or "genetic contamination" of native populations. Second, severe population bottlenecks in introduced populations can reduce restoration effectiveness. Despite the potential for re-introduction efforts to greatly expand brook trout distributions within their native range, there are very few published studies that have documented the effectiveness of this strategy. Given the current lack of published data, further research is needed on the long-term success and potential impacts associated with brook trout re-introductions.

Control of fishing related mortality. Mortality associated with recreational and sustenance fishing can contribute substantially to the dynamics of sport fish populations. Common methods for reducing fishing mortality include catch-and-release regulations and restrictions on fishing tackle. Unfortunately very little is known about the effectiveness of fishing regulations in restoring brook trout populations. Marschall and Crowder (1996) used population models to demonstrate that even under strong harvest-induced mortality of larger fish, brook trout populations were unlikely to be extirpated from a particular stream. It is generally believed that brook trout are relatively resistant to fishing related impacts due to their short life cycle and density-dependent survival and growth of brook trout populations (Grossman et al. 2010). Populations that exhibit strong density-dependence are able to compensate for fishing related mortality through reduced natural mortality and increased growth rates and fecundity. However, to our knowledge there are no published studies that have documented compensatory dynamics in brook trout populations. The effects of harvest in combination with other factors (i.e. physical and chemical degradation, stochastic events) most likely result in population declines that are far more drastic than projected by such deterministic population models. Also, individual brook trout that exhibit more fluvial life history traits (e.g., highly mobile, delayed maturity, larger size, longer-lived) may be highly susceptible to angler harvest, and loss of these individuals from a population may have a disproportionate effect on overall population productivity.

Stream channel restoration. Stream channel reconstruction and structural habitat enhancement are often employed in an attempt to restore degraded riverine ecosystems. Under this restoration strategy, natural (e.g., large woody debris) and artificial structures (e.g., cross-vanes and j-hooks) are placed within the stream channel to compensate for the loss of habitat complexity caused by past land-use practices, such as livestock grazing and channelization. In terms of salmonid restoration, stream channel and physical habitat restoration (i.e. in-stream placement of large woody debris) has been shown to result in an increase in local density and biomass (numerous studies summarized in Roni et al. 2002). However, there are several caveats to the many success stories. First, degraded physical habitat is often accompanied by other impacts such as degraded water quality. Therefore, restoration actions targeted at improving physical structure will only be effective if no other limitations are present. Second, many structural restoration actions take place on relatively short stream segments (< 1 km). Such restoration actions may have little effect on salmonid populations that operate at larger spatial scales (Roni et al. 2010). Third, there has been an alarming lack of monitoring following such restoration efforts, and consequently, it is unclear whether the observed positive response by trout to stream habitat enhancement persists over time. One major concern associated with stream channel restoration is that we do not know if these projects will remain stable in response to major flood events. Also, habitat enhancement projects have a tendency to attract both fish and fishermen, which can increase the overall level of mortality in a population, despite improved habitat conditions.

Limestone sand application. Limestone has increasingly been used to mitigate the effects of stream acidification resulting from acid precipitation and acid mine drainage worldwide, and this is particularly true for Appalachian brook trout streams (Figure 4). Typically, limestone sand with high CaCO₃ content is deposited in the headwaters of areas impacted by acid deposition, neutralizing acid as it is moved downstream within the bed load. Acid remediation has been shown to be very effective at restoring many ecosystem characteristics, such as water chemistry parameters that limit brook trout survival (i.e. pH and alkalinity) (McClurg et al. 2007). As a result, brook trout biomass, density, and reproductive success are most often enhanced or completely restored in treated streams (McClurg et al. 2007; Clayton et al. 1998). Despite its effectiveness in brook trout restoration, acid remediation through the application of limestone sand only treats the symptoms of stream acidification and not the source. Thus, treatment must be continued indefinitely in order to be effective. In addition, as with stream habitat enhancement, most acid remediation projects are conducted at a relatively small scale. Recent research indicates that full system recovery is limited because treated streams remain isolated within acidic networks (McClurg et al. 2007). Consequently, the most effective remediation strategies will require treatment of interconnected drainage networks rather than isolated stream reaches (Petty and Thorne 2005).

Riparian zone management. Many factors that limit brook trout survival are intimately tied to the riparian zone — the ecosystem located at the interchange between streams and their surrounding watershed. Riparian zones are often degraded as a result of anthropogenic disturbances such as forestry, agriculture and grazing, and development. Intact riparian zones are critical to stream ecosystems because they control the supply of sediment, nutrients, food (i.e. insects and coarse particulate organic matter) and large woody debris (i.e. in-stream fish habitat) to receiving streams. Furthermore, wooded riparian areas help shade the stream channel, drastically decreasing in-stream summer temperatures. Thus, riparian zone management typically involves replanting and protection through the use of best management practices (e.g. forestry buffers and cattle fences) (Roni et al. 2002). Not surprisingly, trout density, biomass, and reproductive potential have been shown to increase in response to increased riparian condition following restoration efforts (Carline and Walsh 2007; Summers et al. 2008). However, the extent to which riparian zone management will succeed in the face of current climate change and expanding human population is unknown. This approach, because it represents a sustainable, long-term solution to stream ecosystem improvement, may be one of the most important brook trout restoration tools available to conservation scientists moving forward.

Dispersal barrier removal. Habitat fragmentation or loss of stream connectivity within stream networks has been identified as a significant factor contributing to fish population declines (Roni et al. 2002) (Figure 5). This is true for brook trout in central Appalachia that often disperse to seasonally important habitats to maximize reproductive and foraging success (Petty et al. 2005). The steep mountainous topography characteristic of the brook trout's native range has led to the construction of a vast number of culverts associated with small road crossings (Roni et al. 2002), many of which are impassible by brook trout (Poplar-Jeffers et al. 2009). Culvert replacement projects in the western United States have resulted in large increases in spawning and production of various salmonid species (see Roni et al. 2002). Although comprehensive assessments of the benefits of culvert replacement to brook trout populations have not been conducted, strategic culvert replacement has the potential to drastically increase brook trout production throughout its native range. In essence, culvert replacement is a form of large scale habitat creation, especially for species that preferentially spawn in headwater streams, such as brook trout (Poplar-Jeffers et al. 2009). As with species re-introductions, however, culvert replacement would be expected to be of limited value if water quality or structural habitat is deficient.

Exotic trout removal. Throughout much of the southern Appalachian region, non-native brown and rainbow trout have been identified as important contributors to brook trout declines. Rainbows can encroach on brook trout streams, leaving many southern Appalachian brook trout populations confined to extreme headwater reaches isolated above dispersal barriers, such as waterfalls. Increasingly, and in contradiction with the previous approach, dispersal barriers are being constructed to halt the upstream spread of non-native trout. Dispersal barrier construction typically is coupled with exotic trout removal to ensure that brook trout populations can remain isolated from competition with brown and rainbow trout. It is unclear whether or not this is helpful, though there has been evidence of success from cutthroat trout (Shepard et al. 2002). There are obvious problems with this approach. Most notable, is the fact that isolation of brook trout from a regional network can reduce their overall productivity and make the populations more susceptible to local extirpation as a result of extreme environmental events, such as floods or droughts (Kruse and Hubert 2004).

At the present time, it is safe to conclude that our effectiveness in restoring brook trout populations in the central and southern Appalachian region is highly variable. It is likely that current approaches are limited by an incomplete understanding of: 1 — factors that limit brook trout populations within different parts of their range; 2 — the proper spatial scale at which brook trout populations should be managed; 3 — how combinations of restoration actions can be used to address multiple limiting factors; and 4 — how traditional restoration actions may be influenced by climate change. Consequently, we propose the following plan of action for improving the science and application of brook trout restoration.

Limiting Factors and Brook Trout Population Dynamics at a Watershed Scale. Historically, most studies of brook trout population ecology have focused on isolated populations within relatively small headwater streams. There is emerging evidence, however, that brook trout population dynamics may be influenced by factors and processes operating at the drainage network scale (Petty et al. 2005, 2012). We have found that the factors limiting brook trout populations vary depending on the location of the focal population within the drainage network. For example, headwater stream populations may be strongly limited by acidification and reduced reproductive success. However, populations in medium sized streams are dependent on immigration from smaller streams, and consequently, may be limited by the productivity of nearby streams and by dispersal barriers. Finally, brook trout populations in larger rivers are fully dependent on highly mobile individuals that move between headwater spawning habitats and larger river foraging habitat (Figure 6). These populations can be strongly limited by poor habitat, high water temperatures, angler harvest, and dispersal barriers. If brook trout populations are influenced by watershed scale processes rather than localized stream segment scale processes, then it is likely that: 1 — factors limiting populations in one area will affect populations in another area; 2 — the overall metapopulation will be limited by multiple rather than single factors; and 3 — connectivity among subpopulations will be more important to population viability than local habitat conditions. Unfortunately, the degree to which brook trout populations are influenced by network scale processes throughout their range is largely unknown, and this will limit the success of restoration.

Integrating Multiple Restoration Actions to Achieve Watershed Scale Restoration Goals. A major finding from surveys of stream restoration projects is that most projects represent a single restoration action implemented at a localized spatial scale. For example, some of the most popular (and most expensive) projects involve construction of stream habitat enhancement structures along a 1–2 km corridor. However, if brook trout populations are limited by multiple factors operating across multiple spatial scales, then localized, single-factor restoration actions will likely produce minimal benefits. Consequently, we propose a process that combines multiple restoration actions that address multiple limiting factors across multiple locations within a drainage network. For example, in the upper Shavers Fork, we have designed a brook trout restoration strategy that combines: 1 — acid remediation, 2 — culvert replacement, 3 — in-stream and riparian habitat enhancement, and 4 — angler harvest restrictions in an effort to address multiple limiting factors operating in different areas of the watershed. Because acidification is the dominant factor limiting brook trout within the watershed, limestone sand additions in targeted headwater streams is an essential first step. However, the overall effectiveness of acid remediation will likely be limited unless we address factors currently limiting brook trout dispersal, growth, and survivorship within the watershed. Culvert replacement is needed in several areas to enable brook trout to move freely between headwater spawning areas and larger mainstem feeding areas. In-stream and riparian habitat enhancement along the mainstem (e.g., pool creation and increased structural complexity through large woody debris additions) is needed to provide high quality foraging microhabitats as well as thermal refugia (Petty et al. 2012). Finally, harvest restrictions are needed to protect trout that may be attracted to the habitat enhancement structures. We believe that integrated approaches such as the one described for Shavers Fork, WV would facilitate brook trout restoration in other portions of their range. The specific details of which approaches need to be addressed first and where would, of course, vary depending on the specific location of interest.

Climate change considerations. Finally, all restoration efforts must be considered within the context of climate change. Brook trout are expected to be highly susceptible to hydrologic and temperature changes. For example, under various climate change scenarios, Flebbe et al. (2006) calculated potential reductions in brook trout habitat between 53 and 97%. Potential changes in brook trout habitat suitability as a result of climate change have important implications for brook trout restoration now. We do not want to spend millions of dollars on habitat enhancement in rivers that are likely to become too warm to be inhabited by brook trout. Nor do we want to focus on replacing culverts on streams that will likely become isolated over time due to downstream warming. Consequently, as we develop integrated watershed restoration plans for brook trout like the one described for Shavers Fork, it is essential that we have tools for assessing the potential impact of climate change on the effectiveness of the restoration plan. In addition, it may be a good idea to begin considering how present-day restoration actions may be used as protective measures for brook trout in the future. For example, acid remediation and culvert replacement projects completed now could be planned within the context that the restored streams may provide thermal refugia for brook trout in the future.

In conclusion, numerous approaches are being used to aid in the recovery of brook trout populations throughout their native range. However, many projects are not effectively monitored, and consequently there is a general lack of information on the overall effectiveness of restoration efforts. An additional shortcoming of brook trout restoration in general is a failure to address multiple factors that may limit brook trout populations at a watershed scale. We propose a process for improving brook trout restoration that includes: 1 — additional foundational research on factors limiting brook trout populations at multiple scales; 2 — development of restoration plans that integrate multiple approaches to address multiple limiting factors and increase population connectivity within whole watersheds; and 3 — development of analytical tools that can be used to plan brook trout restoration strategies within the context of climate change.