THE FIRST TWO CHAPTERS in Part 2 examined how fish species and assemblages are affected by broadscale landscape features, how various models relate assemblages to the environmental variables, how fish assemblages are formed, and the role that movement plays over different life-history stages in allowing fishes to access new habitats and to move among habitats so that their fitness is maximized. This chapter focuses primarily on the temporal and spatial dynamics of fish assemblages, or how fish populations and assemblages cope with relatively short-term physical and biotic challenges.

Understanding the type, frequency, and magnitude of variability in fish assemblages is important for several reasons (e.g., Grossman et al. 1990; Matthews 1998). First, assessing the impact of anthropogenic environmental changes requires knowing the background level of natural variation in assemblages. Second, the degree to which assemblages are resistant to changes over space and time is related to the strengths of control mechanisms operative within the assemblage. Although assemblages generally are structured and not random collections of species from a regional species pool (Chapter 5), once established, they may be acted upon by external or internal processes. With some exceptions (Strong 1983), assemblages showing high variation in species composition and abundances may primarily be governed by external, stochastic (i.e., random) processes such as floods, droughts, or other major events. These events can control such processes as species persistence, colonizations, or even extinctions. In communities with strong stochastic influences, the importance of biotic interactions (i.e., competition or predation) in affecting community structure is considered to be lessened because of the frequent changes in species composition. In contrast, assemblages that show little variation may be controlled primarily by deterministic processes, such that the characteristics of the environment result in a particular suite of species (e.g., the landscape filters described in Chapter 4). In assemblages that show little variation in species composition, the possibility of well-developed biotic interactions is considered to be greater (Grossman et al. 1982; Lepori and Malmqvist 2009). Importantly, processes controlling communities should not be viewed in an either-or situation. Stochastic and deterministic processes can act hierarchically (i.e., stochastic processes influence the species on which deterministic processes act). The relative importance of stochastic versus deterministic controls varies with disturbance levels, although not necessarily monotonically (Lepori and Malmqvist 2009).

RESPONSES TO ENVIRONMENTAL PERTURBATIONS

Types of Perturbations

Natural perturbations have shaped the evolution of fish populations and, in the case of severe events, have resulted in the local extirpation of populations or the total extinction of species. For instance, large-scale Cenozoic climatic changes resulted in the extinction of numerous western North American fishes at the end of the Miocene and also the early Pleistocene (G. R. Smith 1981). Natural perturbations include droughts, floods, fires within the watershed, climatic changes, and biotic changes (such as the addition or loss of a predator). Human-induced changes might include chemical spills or piscicide applications; changes in land use, such as mining, agriculture, or timber harvesting resulting in flooding, increased water temperature, nutrient or herbicide runoff, and erosion; major barriers to fish movement as a result of dams or water diversions; stream channelization; and the introduction of nonindigenous species.

One way to view both natural and human-caused disturbances is by their extent. Events that persist longer than the life spans of the species in an assemblage and impact large spatial areas are referred to as press disturbances, in contrast to pulse disturbances, which are of short duration and are generally point source or brief hydrologic events (Bender et al. 1984; Detenbeck et al. 1992). Based on Detenbeck et al. (1992), press disturbances would include impacts of channelization, large-scale habitat alterations, timber harvesting, mining, and changes in nutrient input; pulse disturbances would include floods, chemical spills, droughts, nonchemical removal of biota, and localized construction activity.

Determining what amount of environmental variation actually represents a disturbance or perturbation (since the terms generally are used interchangeably although there are exceptions; e.g., Pickett and White 1985) to aquatic organisms is also challenging—especially for terrestrial, hominid biologists! Natural variations in physical conditions, even some viewed as “a disturbance,” are generally beneficial in the longterm to the well-being of aquatic systems. This would include changes in stream flow (including flow into lakes, ponds, and reservoirs), turbidity, temperature, ice cover, or insolation. For instance, without periodic high, scouring flows in streams, streambed complexity (Mount 1995) and complexity of riverine food webs (Wootton et al. 1996; Power et al. 2008) can be greatly reduced, resulting in population declines or loss of fish species. Likewise, the annual or semiannual turnover in many lakes results in redistribution of nutrients to surface waters and oxygenation of bottom waters (Wetzel 2001).

The recognition of the value of periodic disturbance in ecological communities in the 1970s and 1980s led to models of how periodic disturbance fostered increased species diversity. This corresponded with the recognition that most communities probably did not exist at some sort of steady state or equilibrium (Levin and Paine 1974; Sousa 1984). The intermediate disturbance hypothesis (Levin and Paine 1974; Connell 1978) predicts that the greatest species richness would occur at some intermediate level (intensity and/or frequency) of disturbance. The logic is basically that intermediate levels of disturbances provide sufficient time for species to colonize affected patches of habitat yet keep the habitat from being dominated by only a few species (Connell 1978; Sousa 1984). In a similar way, the dynamic equilibrium model (Huston 1979) predicts that diversity of communities is the outcome of two processes—the rate of population growth of competing species, balanced against the frequency of population reductions, caused by various types of disturbances. In contrast to disturbance functioning by mediating competitive interactions between species, the role of intermediate disturbance in a study of stream macroinvertebrates was due to the removal of more sensitive species, so that invertebrate communities converged to a core group of species moderately resistant to disturbance (Lepori and Malmqvist 2009).

What constitutes a disturbance also changes over ecological and evolutionary time and among taxa. Viewed in the evolutionary context of species and assemblages, a force that once was a major disturbance might be less so today given the strong selection for populations to withstand environmental change (Sousa 1984). In ecological time, as elaborated later in the chapter, life-history stages of a species differ in their abilities to respond to disturbances, just as species differ in their responses. Furthermore, the seasonal timing of disturbance can affect the level of impact on fish populations and, because of the wide variation in body size and mobility among fish species, disturbance must be viewed relative to the spatial and temporal dynamics of species (Pickett and White 1985).

Efforts to define disturbance have taken two main approaches. One approach defines a disturbance by its magnitude, whereas the other defines a disturbance by the population, species, or community responses to it or to its impact on the physical environment (Resh et al. 1988; Matthews 1998). In the former case, a sudden change in water temperature or stream flow that exceeded some arbitrary value, say ± two standard deviations, would be judged as a disturbance, whereas a change of less than ± two standard deviations would not. In the latter case, if there were no apparent biological or physical response to what would seem to be a disturbance, such as a major flood event, then the event would not be considered a disturbance. The latter approach has generally been preferred (e.g., Resh et al. 1988), and a useful working definition of a disturbance proposed by White and Pickett (1985) and used by other authors (e.g., Resh et al. 1988; Yount and Niemi 1990) is “any relatively discrete event in time that disrupts ecosystem, community, or population structure, and that changes resources, availability of substratum, or the physical environment.” As such, a disturbance is “the primary event, or cause, from which certain effects follow” (Yount and Niemi 1990).

The Metric

Responses to environmental change can basically be measured by the presence or absence of species, irrespective of the actual numbers or relative abundance of individuals. This qualitative measure is referred to as persistence, in contrast to stability, which is based on abundance measures (Connell and Sousa 1983). Quantitative measures include relative abundances, or actual numbers or densities of the component species. The choice of metric has a strong influence on the detection of change, or lack thereof, in fish populations (Rahel et al. 1984; Yant et al. 1984; Matthews et al. 1988; Grossman et al. 1990; Rahel 1990; Matthews 1998). For instance, presence-absence, ranks in abundance, relative abundance measures, and actual numbers of individuals form a transformation series of increasing sensitivity to change. That numbers of individuals of a given species show the greatest variation is not surprising, especially because most long-term studies of fish assemblages employ sampling techniques that are not designed to provide rigorous quantitative data on population sizes (Matthews 1998).

Spatial and Temporal Scales

In addition to the appropriate metric, the spatial and temporal scales over which a measurement is made also affect the outcome. To assess stability, the temporal scale must encompass at least one full turnover in the assemblage (Connell and Sousa 1983); if it does not, then what is really being measured is simply the impact of long-lived organisms on the local community. This point can have a major impact on apparent regional differences in responses of fish assemblages to environmental change. Some southwestern fish assemblages, such as in the San Juan River of the Colorado River drainage, consist primarily of species like Flannelmouth (Catostomus latipinnis), Bluehead (C. discobolus), and Razorback (Xyrauchen texanus) suckers; Roundtail Chub (Gila robusta); Speckled Dace (Rhinichthys osculus); and Colorado Pikeminnow (Ptychocheilus lucius) (Tyus et al. 1982; Propst and Gido 2004). With the exception of the short-lived (ca. 3 years) Speckled Dace, these San Juan River species commonly live more than 20 years, and in the case of Colorado Pikeminnow and Razorback Sucker, over 40 years (John 1964; McCarthy and Minckley 1987; Scoppettone 1988; Lanigan and Tyus 1989; Osmundson et al. 1997). In contrast, southeastern fish assemblages, such as in Black Creek of the Pascagoula River drainage, Mississippi (Baker and Ross 1981; Ross et al. 1987), are composed primarily of small minnows, topminnows, darters, and sunfishes, most of which have life spans of only 1–5 years (Ross 2001). A study of 4–5 years would essentially capture one complete assemblage turnover for the Black Creek fishes, whereas an equivalent study in the San Juan River would need to extend to 20 years or more to achieve the same result. In probably the majority of studies, the temporal scale is defined more by the duration of funding or graduate student tenure (both commonly on the order of 1–5 years) than by consideration of the life history of the fishes—with some notable exceptions

The spatial scale of a study also has a major impact on the ultimate outcome (Connell and Sousa 1983; Rahel 1990). If the spatial scale does not include the normal population bounds of the component species (see Chapter 5), then it is likely that any measure will record extensive changes in assemblage structure. In contrast, a large study area might include a number of subpopulations comprising various metapopulations of the component species (see Chapter 4), so that variation or loss of taxa in one area is damped out by their survival in another. Connell and Sousa (1983) suggest that the spatial scale should correspond to the least area that is necessary for the recruitment of adults through successful reproduction, survival, and growth of young. Recalling the types and extent of movement in Chapter 5, this guideline would result in widely differing spatial scales, depending on the species and region. However, again with a few notable exceptions, the spatial scale of most studies is somewhat arbitrary or driven by sampling logistics or cost. Thus not only the analytical scale, as illustrated previously, but also the temporal and spatial scales of the analysis have strong effects on the outcome and, not surprisingly, because of the interactions between metrics and scales, there are conflicting views on the nature of change in fish populations and assemblages.

Assessing Assemblage Change

The stability and persistence of assemblages should be investigated on multiple levels within the hierarchical framework (Rahel 1990). Separating actual changes in species presence or absence from artifacts of sampling also is a pervasive and significant problem (Magnuson et al. 1994). Preferably, the goal of current research should not be to determine whether local assemblages change or not—virtually all local assemblages undergo change as individuals are added or removed (due to natality and mortality, or movements). Instead, as the temporal extent of data increases (several to many samples over a period of years to decades), and as change is measured on multiple spatial scales, it becomes of more interest to ask how much a local fish assemblage, or distribution of fish species within a watershed, have changed during various intervals (Ross and Matthews, in press). Also, an important issue is whether changes largely are driven by major events such as extensive droughts or floods, with little change during times that lack apparent disturbance factors. In other words, are the changes that can be observed in fish assemblages over long periods of time related more to gradual changes, or to dramatic events that may (or may not) leave their mark for years or more?

Dealing with Environmental Change

Fish populations can deal with environmental or biotic stressors in three primary ways. First, populations might lack means of dealing with perturbations and be eliminated from a region altogether. Second, individuals in a population might show resistance by withstanding such challenges through morphological, physiological, or behavioral adaptations, such as refuge-seeking behavior that would overall increase their tolerance to environmental perturbations. Finally, populations might emigrate from, or perish in, the stressed habitat but recover following a perturbation by return immigration of the displaced individuals or colonization by individuals from other populations, such that pre- and postdisturbance assemblage structures are the same or similar. This approach to dealing with environmental change was termed adjustment stability by Connell and Sousa (1983) and resilience by Dodds et al. (2004). Connell and Sousa (1983) considered that this response included two components: amplitude and elasticity. Amplitude is a measure of how far a population or assemblage can be displaced from its predisturbance state and still return; elasticity, drawing further on the analogy with a rubber band, is a measure of how quickly populations or assemblages can return to a predisturbance condition.

Resistance

It is not surprising that fish assemblages occurring in geographic regions prone to extreme climatic conditions are generally persistent in the face of such environmental challenges, given that these faunas have been under strong, long-term selection to deal with such conditions, and intolerant species would have been extirpated (C. L. Smith and Powell 1971). Examples of fish faunas inhabiting areas prone to disturbance include faunas of Great Plains streams. Fishes in these streams are subjected to periods of low flow and even dewatering as well as to intense periods of flooding. Such conditions likely occurred before widespread human-caused changes in the nineteenth and twentieth centuries, although the amount of silt has probably increased as a consequence of changes in land use (Matthews 1988).

In response to this environment, some fishes have evolved increased tolerances to low dissolved oxygen levels and high temperatures. Comparisons of physiological tolerances among minnow species from more benign upland streams in Arkansas with those inhabiting apparently harsher Great Plains streams in central and western Oklahoma generally showed that minnows from harsh environments were more resistant (Matthews 1987). As a group, Matthews showed that the minnows from the harsh streams had significantly greater tolerance to high water temperatures compared to fishes from the relatively benign streams, although one prairie minnow, the Emerald Shiner (Notropis atherinoides), had a critical thermal maximum (CTM) more in line with the upland fishes (Figure 6.1). However, Emerald Shiners, along with three other plains minnows, showed better survival at low oxygen levels compared to upland fishes or Blacktail Shiner (Figure 6.1).

FIGURE 6.1. Contrasts in physiological resistance of minnows from relatively harsh versus benign environments. Harsh environments include prairie streams from central and western Oklahoma; benign environments include streams from upland regions of Arkansas. Oxygen tolerance was measured as the percentage of fish surviving 8.5–10 h at low oxygen levels (0.2=0.9 ppm dissolved oxygen). Based on data from Matthews (1987).

Pupfishes (genus Cyprinodon) occur widely in fresh and brackish water habitats in Mexico (Miller 2005), in coastal brackish water areas along the Gulf of Mexico and Atlantic coasts (Johnson 1980; Nordlie 2003), and in desert regions of the southwestern United States (Naiman and Soltz 1981). In all regions, Cyprinodon species show high resistance to temperature extremes (Feldmeth 1981; Bennett and Beitinger 1997; Nordlie 2003). For instance, the Sheepshead Minnow (Cyprinodon variegatus) of the Atlantic and Gulf coasts can survive temperatures from a low of −1.8° C to a high of 43° C, the widest temperature range of any of the over 200 estuarine/salt marsh fishes reported by Nordlie (2003). Populations of cyprinodont fishes inhabiting the Death Valley region of Nevada and California also have wide temperature tolerances, being able to withstand temperatures of < 1° C to 40–44° C (Brown and Feldmeth 1971; Soltz and Naiman 1978; Feldmeth 1981).

Refuge-seeking behavior also allows fishes to resist adverse conditions in their environment and can have a role in surviving floods and droughts. Fishes in a variety of regions increase their resistance to downstream displacement during floods, especially during the winter when lowered water temperature limits their swimming ability, by actively selecting habitats with large structure such as woody debris, rocks, or other large and relatively immovable structures. Cutthroat Trout (Oncorhynchus clarkii) in a tributary of the Smith River in California showed twice the site fidelity in pools with large woody debris compared to those without (Harvey et al. 1999). During a winter flood event, trout in pools with large woody debris tended to remain in those pools in contrast to the greater movement shown by trout in less complex habitats. Fishes in an arid eastern Oregon stream also were more resistant to floods in structurally complex habitats compared to simple habitats (Pearsons et al. 1992). In the southeastern United States, Bayou Darters (Nothonotus rubrum), a species endemic to the Bayou Pierre system of Mississippi, responded to winter water temperatures (7.5–11° C) in an artificial stream by shifting their distribution to habitat patches with larger particle sizes of coarse gravel and pebbles (Figure 6.2A). In addition, under winter conditions, as flow increased from 14 to 35 cms-1, Bayou Darters selected habitat patches with large refuges to current (in this case bricks 14 × 7 × 7.5 cm) over habitats with just pebbles (Figure 6.2B).

FIGURE 6.2. Behavioral resistance of Bayou Darters (Nothonotus rubrum) to high winter stream flow in western Mississippi.

A. At cold temperatures (7.5=11.0° C) in a laboratory stream with a current speed averaging 31 cms-1, Bayou Darters shift their habitat selection to patches with larger particle sizes compared to habitat selection at warm temperatures (22° C).

B. Even in habitats with large substrata, Bayou Darters select patches with larger refuges (bricks) at moderate versus low current speeds. In both figures, bars are 95% confidence intervals. Based on data from Ross et al. (1992b).

FIGURE 6.3. Behavioral resistance of adult and one-day-old Sonoran Topminnow (Poeciliopsis occidentalis) to displacement by floods, compared to the nonnative Western Mosquitofish (Gambusia affinis). Data show mean number retained in a laboratory stream system for each of ten, 60 s trials with a 60 s rest period between each trial; six fish of each species were used in each trial. One-day-old Western Mosquitofish only survived for five replicates. Based on data from Meffe (1984). The shaded area of the map inset shows the approximate boundaries of the Sonoran Desert.

Sonoran Topminnow (Poeciliopsis occidentalis), a small poeciliid native to streams, marshes, and springs in the Sonoran Desert, has evolved behavioral responses to flash floods that commonly occur in the region, in contrast to the morphologically similar Western Mosquitofish (Gambusia affinis), which is not native to the region (Figure 6.3). In habitats that are not periodically disturbed, Western Mosquitofish can reduce or eliminate Sonoran Topminnow, probably through predation on the young (Meffe 1984). Using field observations in a tributary of the Santa Cruz River in southern Arizona, combined with laboratory experiments, Meffe (1984) showed that Sonoran Topminnow of all life-history stages responded to floods by rapidly moving to shoreline eddies and remaining there until high flows receded. In contrast, Western Mosquitofish responded more slowly and in a less organized manner to flooding and tended to move back out into high flows sooner, thus exposing themselves to downstream displacement. As a consequence, Sonoran Topminnow showed stronger resistance to downstream displacement by flooding in contrast to the nonnative Western Mosquitofish, although both species showed improvements in flood resistance through repeated exposure (Figure 6.3). By testing one-day-old fish, Meffe also demonstrated that the behavioral response to floods is innate in Sonoran Topminnow; one-day-old Western Mosquitofish were almost all displaced. Western Mosquitofish were also completely displaced from Sabino Creek, another southern Arizona stream, by a record winter flood, whereas a native minnow (Gila Chub, Gila intermedia) and a nonnative centrarchid (Green Sunfish, Lepomis cyanellus) were not (Dudley and Matter 1999).

Morphology, in concert with appropriate behavior, can also provide resistance to harsh environments. For example, most all fishes that obtain oxygen from the water will move closer to the water’s surface as oxygen levels are depleted—the response is termed aquatic surface respiration (ASR) (Kramer 1987). However, most fishes, because of jaw morphology and head shape, must spend additional energy through body or fin movements to maintain the extreme body angle required for ASR and cannot survive severe subsurface oxygen depletion for extended periods of time (Lewis 1970). Fishes such as livebearers and topminnows have flattened heads and superior mouths, morphologies that are particularly adapted to ASR, and are able to use the highly oxygenated surface film while remaining in a nearly horizontal (< 10°) position. By exploiting the surface film, these fishes can survive extended periods in water that is otherwise low in oxygen (Lewis 1970; Kramer 1987; Timmerman and Chapman 2004). More recent work indicates that ASR might be most important as an immediate response measure to low oxygen levels. For instance, Sailfin Mollies (Poecilia latipinna), as well as a variety of other fish groups, are able to gradually increase oxygen capacity of the blood when chronically subjected to an oxygenpoor environment (Timmerman and Chapman 2004). This occurs through an increase in red blood cells and through increased hemoglobin concentration.

Resilience

Fish populations are often faced with environmental perturbations that are too severe for one or all life-history stages to resist through morphological, physiological, or behavioral mechanisms. Such perturbations might include extreme floods, drying of essential habitats such as feeding or spawning areas, changes in water quality, or total drying of an aquatic habitat. Although the initial effect can be the total or partial loss of species making up an assemblage or the lack of successful reproduction, the ability to recover once environmental conditions become more favorable is described by resilience.

Resilience to major environmental perturbations is provided through the ability of fish populations and assemblages to repopulate an area once conditions improve. This may occur through the return of displaced individuals and through the often-accelerated production of new individuals (Ross et al. 1985; Matthews 1986b; Fausch and Bramblett 1991). The extent of resilience is influenced by the size of the affected habitat and by the size and proximity of refuges where fishes can survive. Watershed geometry (e.g., Chapter 4; Figure 4.3) plays an important role in resiliency (Grant et al. 2007).

In southern Oklahoma, fishes in Brier Creek (see Chapter 5; Figure 5.4) recolonized a dewatered section of stream within four months once flow resumed. Recolonization to this pulse disturbance was initially by movement of fish out of isolated pool refugia, followed by spawning. In a southeastern study, Albanese et al. (2009) removed adult and juvenile fishes from 416 m and 426 m reaches of two small streams, Middle Creek and Dicks Creek, located in the James River drainage of Virginia (Figure 6.4). They followed recolonization for approximately one year along a 130 m reach in Dicks Creek, and two years along a 126 m reach in Middle Creek. The reaches were located in the middle of each of the two removal sections (Figure 6.4). The larger Dicks Creek site involved 19 species, whereas the smaller Middle Creek site involved 6 species. The fish fauna in both streams was dominated by minnows, which make up 85% of the fish in Dicks Creek and 92% in Middle Creek.

The resilience of individual species studied by Albanese et al. (2009), measured by their rates of recovery to the simulated pulse disturbance, varied widely; rates of recovery also differed between the two streams. Mountain Redbelly Dace (Chrosomus oreas) rapidly recolonized, reaching over 60% of the original population size within one month in Dicks Creek but less than 20% in Middle Creek (Figure 6.4). At the end of one year, Mountain Redbelly Dace populations had fully recovered in Dicks Creek, but required an additional year for full recovery in Middle Creek. After one year, five of the eight censused species in Dicks Creek attained 80% or greater recovery; Shadow Bass (Ambloplites ariommus) and Blacknose Dace (Rhinichthys atratulus) showed much lower resilience. In Middle Creek, three of the five censused species reached 90% or greater recovery after two years, whereas Torrent Sucker (Thoburnia rhothoeca) did not recover and Rosyside Dace (Clinostomus funduloides) only reached 60% of the original population size. Even though species varied significantly in their resilience to the defaunation, the fish assemblages, as measured by pre- and postremoval similarity, appeared resilient. This occurred because abundant species remained abundant, and more abundant species have a greater effect on faunal similarity measures that include relative abundance (Matthews 1998). As stressed by Albanese et al. (2009), this has important conservation implications. Measures that focus only on the assemblage level could overlook the loss of rare species following natural or human-caused perturbations.

FIGURE 6.4. Varying levels of resilience of two southeastern fish assemblages as demonstrated by recolonization of experimentally defaunated areas. The map shows the study area, which was located in the James River drainage of Virginia; ovals are impoundments. Recovery was followed for approximately one year in Dicks Creek and two years in Middle Creek. Based on data from Albanese et al. (2009).

In western North America in the Willamette River drainage of western Oregon, Lambertiet al. (1991) followed recovery of Cutthroat Trout in Quartz Creek, a high-gradient stream that had suffered a pulse disturbance in the form of a catastrophic debris flow. The debris flow had severe impacts on the physical and biotic characteristics of a 500 m stream reach. Physical changes included loss of woody debris, loss of canopy cover, a reworking of channel sediments, and an overall simplification of the channel, resulting in reduced hydraulic retention. Chlorophyll α was low immediately after the debris flow, but the newly opened canopy and the reduction in grazing by macroinvertebrates later resulted in a doubling of chlorophyll α compared to a control reach. Macroinvertebrate density initially showed high variation, followed by recovery after one year to densities shown in the control reach, and recovery of species richness after about two years. Cutthroat Trout, the only fish species in Quartz Creek, were initially extirpated. Resilience, as measured by percent recovery to predebris flow conditions, increased rapidly after one year; overall recovery of Cutthroat Trout, which required three years, initially began by immigration of juvenile fish (age-1+) into the disturbed area, followed in the second and third years by enhanced recruitment of fry.

Although responding to press disturbance, salmonids in a Canadian stream required at least a year to successfully recolonize a rewatered section of the Bridge River after a long period of no or greatly reduced water flow. The Bridge River, a British Columbia tributary of the Fraser River, was impounded in 1963 and most of the captured flow (annual mean discharge of 100 m³s⁻¹) was redirected into another watershed for hydropower production (see also Chapter 14). As a consequence, there was no flow in a 4 km section below the dam, after which groundwater and small tributaries resulted in a small flow of 0.7 m³s⁻¹ for the next 11 km before being substantially augmented by a large tributary (Decker et al. 2008; Bradford et al. 2011). Once the flow was restored in the 4 km reach, the flow was only at a level of 2–5 m³s⁻¹, 2–5% of the original annual mean discharge (as the channel was regraded to accommodate the reduced flow), but did result in some positive responses. There was rapid recolonization of periphyton and aquatic insects, but colonization by fishes was much slower. Juvenile salmonids (primarily Steelhead and Rainbow Trout, Oncorhynchus mykiss; Coho Salmon, O. kisutch; and Chinook Salmon, O. tshawytscha) did not move upstream even though an invertebrate prey base was available within three months. Instead, colonization was primarily the result of upstream movement of adult anadromous fishes that successfully spawned in the restored habitat. Coho and Chinook salmon spawned in the fall of 2000 and Steelhead spawned the following year. By one year after the resumption of flow, populations of age-0 Rainbow Trout and juvenile Coho and Chinook salmon in the rewetted area were equivalent to downstream populations in the continuously wetted site.

Resilience is also shown by life-history responses of fishes, such as the timing and duration of reproductive cycles or the length of the reproductive life span. The Split-tail (Pogonichthys macrolepidotus), a cyprinid endemic to the Sacramento-San Joaquin Estuary in west-central California, requires inundated floodplains for successful reproduction (Sommer et al. 1997). Submerged terrestrial vegetation on inundated floodplains is used as a feeding area by prespawning adults, as a spawning substratum, and as a larval nursery area. Low-flow years result in substantial reductions in the production of age-0 fish, in contrast to large increases of age-0 fish during wet years. Because the adults have a reproductive life span of three or more years, as well as a high fecundity, the populations are moderately resilient to periodic drought years that limit successful reproduction (Sommer et al. 1997).

Fishes in a small Ontario, Canada, lake also demonstrate resilience to harsh conditions through survival of long-lived adults. The lake was acidified by the addition of sulfuric acid for eight years and then recovery studied for 13 years as part of a large investigation on the effects of acid precipitation (Mills et al. 1987). Three of the five species studied by Mills et al. (1987) (Lake Trout, Salvelinus namaycush; Pearl Dace, Margariscus margarita; White Sucker, Catostomus commersonii) survived the acidification but were not able to successfully reproduce as the pH level dropped. Once acidification stopped and the pH began to gradually rise, recruitment of all three species gradually resumed. Two other species, Fathead Minnow (Pimephales promelas) and Slimy Sculpin (Cottus cognatus), were extirpated from the lake during the acidification period, but only Fathead Minnow successfully recolonized from a nearby lake during the 13 years following acidification. Population levels of fishes, especially the top predator, Lake Trout, had not reached preacidification levels by the end of the 13-year study of recovery—likely a reflection of the still recovering prey base.

Fishes also may show resilience to unfavorable spawning conditions by having extended reproductive seasons. The Longnose Shiner (Notropis longirostris), a cyprinid found in small, upland streams in the southeastern United States, has a short life span of only 1–2 years. However, resilience to poor spawning conditions is achieved by having a protracted spawning season that begins in February and can extend into October (Heins and Clemmer 1976; Ross 2001). Similar patterns of extended spawning seasons in association with short life spans are shown for numerous other southeastern minnows such as Red Shiner (Cyprinella lutrensis), Blacktail Shiner (C. venusta), and Weed Shiner (Notropis texanus) (Ross 2001).

In summary, resilience in fishes can be achieved by movement of adults or juveniles back into a previously disturbed area. Resilience to poor spawning conditions or unfavorable conditions for larval/juvenile survival occurs through elevated longevity of adults so that they can wait out poor years. On an annual basis, short-lived fishes show resilience to poor spawning conditions by having extended reproductive seasons. Overall, there is considerable variation in the resilience of fish species and fish assemblages to perturbations. Variation occurs across multiple levels including the nature, timing, and severity of the disturbance; the type and location of the aquatic system; species characteristics; and life-history stage (Schlosser 1985; Detenbeck et al. 1992; Albanese et al. 2009).

Levels of Persistence and Stability in Lotic Systems

Considering a wide range of studies, lotic systems tend to show moderate to high levels of persistence and low to moderate levels of stability, with the degree of stability influenced by the metric used to test it. In a survey of 49 primarily North American stream sites that had been subjected to various types of disturbance, Detenbeck et al. (1992) found full or nearcomplete recovery within two years. Analysis of 25 long-term studies (≥ 2 years; median = 11 years; range 2–45 years) designed or amenable to testing assemblage persistence and stability, and including from 3 to 95 species, showed 76% high persistence and 52% high stability in at least one type of measure (Table 6.1). Environmental harshness, especially if the harshness was related to anthropogenic impacts, had a strong effect on assemblage persistence and stability (Figure 6.5A). In systems judged to have low stress, 100% of the assemblages were persistent and 80% stable. In contrast, for systems judged to have moderate or high stress, only 22% were persistent and 11% stable. In systems with obvious human disturbance, only 14% of the assemblages were considered persistent or stable (Figure 6.5B). However, the sample size is too limited to separate the impacts of human versus natural disturbances; of the 10 studies having moderate to high disturbance, only three were disturbed by nonhuman impacts. The data in Table 6.1 are also biased by geographical region; most of the studies were at lower latitudes (mean latitude = 35.7°; range = 31°–42°) and 84% were done east of the continental divide. However, recall the challenges of assessing assemblage persistence and stability in western fish faunas with long-lived species. There is also a bias in stream size as only two studies dealt with large rivers.

FIGURE 6.5. Impacts of environmental stress (A) and the level of human disturbance (B) on the degree of persistence and stability of lotic fish assemblages, and (C) a comparison of assemblage persistence and stability in lotic and lentic systems. Numbers above bars show sample sizes. Based on data from Tables 6.1 and 6.2.

Examples of Persistence and Stability in Lotic Systems

Brier Creek, an Oklahoma tributary of the Red River (now inundated by Lake Texoma), is routinely subjected to extreme conditions and has been particularly well studied. In spite of extreme conditions, including total dewatering of some stream reaches, the fish fauna over an 18-year period showed strong persistence on a stream-wide basis, in that abundant species continued to remain abundant and rare species remained rare, with only a few exceptions. Stability of the Brier Creek fish fauna showed greater variation, as measured by indices of similarity of the sampled fish fauna among years. The fish fauna at individual collection sites (i.e., at the local assemblage level) showed less persistence and stability.

The timing of perturbations can have a major influence on the resultant impacts to aquatic organisms. If flooding in Brier Creek occurs when fish are spawning, there are severe impacts on larval survival. For instance, Harvey (1987) showed that larval cyprinids and centrarchids that were less than 10 mm TL were displaced downstream and killed by a major flood event.

Long-term data also exist for Piney Creek, a permanent upland stream in the Ozarks (Ross et al. 1985; Matthews 1986b; Matthews et al. 1988) that offers a more benign habitat (i.e., no dewatering and less temperature variation). Not surprisingly, Piney Creek fishes also showed high persistence; however, in contrast to Brier Creek, the fish fauna in Piney Creek also showed greater faunal stability, both overall and at the assemblage level. Piney Creek had a severe flood in 1982; however, immediately after the flood there were no major changes in rank abundance of the 10 most abundant species (Matthews 1986b). Less common species did show changes in abundance, so local assemblages were altered immediately postflood. Eight months after the flood, the overall fish fauna and the fauna at individual collecting stations had essentially recovered to preflood conditions, leading Matthews (1986b) to conclude that the Piney Creek fish fauna showed stability and persistence across years and across a range of flow conditions.

Although some studies have shown that fish assemblages rebound rather quickly from flooding, as discussed previously, and as documented also by Taylor et al. (1996) for mainstem and tributary sites in the upper Red River system of Oklahoma, other studies indicate that floods or droughts acted to change or reset assemblage structure. Matthews and Marsh-Matthews (unpublished data) have recently found that two severe droughts resulted in a substantial change in the Brier Creek fauna, which did not recover to its former state until 3–4 years postdrought. Another example of how fish assemblages are affected by perturbations emphasizes the significance of timing of the event. In Coweeta Creek, North Carolina, a severe drought resulted in three distinct assemblages over a 10-year period corresponding to predrought, drought, and postdrought conditions (Table 6.1) (Grossman and Ratajczak 1998; Grossman et al. 1998).

TABLE 6.1 Long-Term (≥ 2 years) Studies of North American Stream Fish Assemblages Organized from Low to High Levels of Stress and from Low to High Latitudes

For a downloadable PDF of all tables, go to ucpress.edu/go/northamericanfishes

TABLE 6.1 (continued)

In the Sierra Nevada mountains of California, a severe spring flood in Martis Creek shifted the assemblage from being dominated by native, spring spawning species, to domination by the nonindigenous, fall spawning Brown Trout (Salmo trutta) (Table 6.1) (Strange et al. 1992). The importance of timing of floods relative to life history is also shown by responses of a northwestern fish assemblage in the John Day drainage, Oregon. Fishes that spawned in late spring and summer, such as Speckled Dace and Bridgelip Sucker (Catostomus columbianus), showed high losses of young-of-year individuals to summer flooding, whereas early spring spawners, such as Rainbow Trout, were more susceptible to spring flooding when the developing embryos were still in the redds (gravel nests) (Pearsons et al. 1992). In addition, losses of fishes due to flooding were greater in stream sections with low habitat complexity, leading Pearsons et al. (1992) to suggest that complex habitats may act as sources of individuals for the colonization of structurally simple habitats.

Levels of Persistence and Stability in Lentic Systems

There are relatively few studies with suitable data for assessing the temporal persistence and stability of lakes and reservoirs (Table 6.2). However, in an analysis of nine long-term studies (median duration = 15 years; range 11–72) of lentic fish assemblages (eight natural lakes and one impoundment), the levels of persistence and stability were essentially the same as those for lotic systems (Figure 6.5C; Table 6.2). Reduced persistence or stability of assemblages tends to occur in altered habitats and/or in habitats impacted by nonnative plants or animals. In contrast to lotic systems, all but one of the lentic studies had suffered moderate or major human impacts, primarily through commercial fishing pressure, the introduction of nonnative plants and animals, and overall urbanization within the watershed.

Examples of Persistence and Stability in Lentic Systems

Numerous lakes exhibit evidence of negative impacts on persistence and stability. Lake Michigan has received considerable study because of heavy fishing pressure and the introduction of nonindigenous species such as Sea Lamprey (Petromyzon marinus), Alewife (Alosa pseudoharengus), Rainbow Smelt (Osmerus mordax), Coho Salmon, Chinook Salmon, Rainbow Trout, Brown Trout (Salmo trutta), and Brook Trout (Salvelinus fontinalis). Many native species such as Lake Trout, Bloater (Coregonus hoyi), and Cisco (C. artedi), have shown substantial declines and/or increases as numbers of nonindigenous fishes have fluctuated (see also Chapter 15). Although overfishing, Sea Lamprey predation on large fishes, and competitive interactions all contributed to the decline of indigenous fish species, another factor has been predation on early life-history stages (Stewart et al. 1981; Eck and Wells 1987; Miller et al. 1989). Major shifts in species composition of a small Michigan lake after the loss and then reintroduction of Largemouth Bass (Micropterus salmoides) have also been observed (Mittelbach et al. 1995). In Lake Mendota, Wisconsin, which has been impacted by extensive shoreline urbanization and introduction of nonnative vegetation, the fish fauna showed both low temporal persistence as well as low stability (Lyons 1989)

Lentic systems showing greater persistence and stability were generally, but not always, less impacted by habitat alteration or introduction of nonnative species (Table 6.2). In six small Michigan lakes, changes in the fish assemblages were generally low over a four-year period, as determined from a measure of community heterogeneity (based on the average percent dissimilarity over all possible pairs of seine sites within lakes) (Benson and Magnuson 1992). Somewhat surprisingly, heterogeneity among seine hauls within a site (33 m of shoreline) was of the same magnitude as heterogeneity among sites, thus suggesting that fishes were responding to small-scale patchiness of the environment.

In a 15-year study of two depauperate Arctic lakes, Johnson (1994) examined long-term stability of Arctic Char (Salvelinus alpinus) populations. Arctic Char was the sole species in one lake and one of two species in the other. After an initial period of moderate (one lake) and high (the other lake) fishing exploitation, the lakes were allowed to recover for several years. In both lakes, age and size structure of Arctic Char returned to the original condition, indicating population stability.

Using long-term data, Gido et al. (2000) examined stability and persistence of a pelagic reservoir fish assemblage over a 43-year period in Lake Texoma, a large impoundment on the Oklahoma-Texas border. Except for the introduction of two species within the study period, Striped Bass (Morone saxatilis) and Threadfin Shad (Dorosoma petenense), the fauna showed persistence and stability as determined from the rank order of species. Numbers of individuals of each species showed greater fluctuations, with coefficients of variation of the 11 most abundant species ranging from 11–108% over the years.

Persistence and Stability Summary

At the scale of the entire fish assemblage and across a wide range of systems, persistence is fairly common and stability somewhat less so, although both can be impacted by the timing, type, and magnitude of perturbations. Persistence and stability in lentic and lotic systems are generally reduced following severe human disturbances (see also Schlosser 1982; Matthews 1998), and such disturbances occur in both types of systems, although they are more common in lentic (89%) compared to lotic studies (28%). At the population level, rare species, and especially those with limited vagility, recover more slowly or not at all from perturbations, as shown by Albanese et al. (2009) for fishes in the James River drainage, Virginia. Although rare species have less of an effect on most measures of assemblage similarity, there is a greater risk of losing such species either from local habitats or system-wide. Because of this, although responses of assemblages to perturbations indicate generally high stability, postimpact fish assemblages (even if judged highly similar to preimpact assemblages by most measures of assemblage structure) might differ in the loss of rare species.

Although there are fewer lentic compared to lotic studies, those in lentic systems tend to compare longer time intervals (median 15 versus 11 years) and half as many species (mean 10, range 1–20, versus mean 21, range 3–95). Compared to lotic studies, lentic studies were also generally at higher latitudes (mean 49, range 34–64, versus mean 36, range 31–42)—regions that typically have lower fish species diversity. Thus our understanding of persistence and stability in streams, reservoirs, and lakes is incomplete because of relatively few studies and biases in geographic location, species richness, and length of comparisons.

Persistence and Stability of Local Associations

The previous sections dealt with levels of change in species assemblages over time periods of two or more years and over moderate to broad spatial scales. Much less is known about how close contacts of species in associations change over time—for instance how long do multispecies groups remain and do they remain together long enough for reciprocal evolutionary responses (coevolution) to occur? It can be difficult to detect association patterns in species of mobile animals. Individuals found in the same sample may or may not have been in close enough contact to have had direct interactions with each other. The capture of individuals of two species in a sample may not equate to their direct interaction because most survey methods for fishes, or other mobile vertebrates, have fuzzy boundaries and may sample different microhabitats (Ross and Matthews, in press). Also, fishes found together in a relatively long reach of stream (e.g., Marsh-Matthews and Matthews 2002) may, especially in highly structured habitats, occupy different pools or riffles and thus do not encounter each other daily. For mobile animals distributed across a heterogeneous landscape, consistent spatial associations among species could exist because some species (or some life stages within a species) select similar microhabitats totally independent of each other (see also Chapter 13) (Chapman and Chapman 1993; Grossman et al. 1998; Wilson 1999).

TABLE 6.2 Long-Term (≥ 2 years) Studies of North American Lake and Reservoir Fish Assemblages Organized from Low to High Levels of Perceived Stress and from Low to High Latitudes

For a downloadable PDF of all tables, go to ucpress.edu/go/northamericanfishes

Matthews and Marsh-Matthews (2006a) provided one of the clearest studies of the longevity of multispecies associations. Based on 19 snorkeling surveys taken over 22 years, they examined persistence of associations of eight taxonomic species (including minnows, a topminnow, sunfishes, and black basses) and 11 “ecospecies” (with the sunfishes separated into piscivorous adults versus insectivorous juveniles) across 14 adjacent pools within a kilometer reach of Brier Creek, Oklahoma (Figure 6.6). For each completed survey of the 14 pools (Figure 6.6A), species associations were compared by constructing a triangular similarity matrix of species pairs based on relative abundances (Figure 6.6B). Next, the strength of species associations over time was determined by sequentially comparing the 18 matrices using the Mantel test, a statistical procedure for comparing the correlation between matrices (Legendre and Legendre 1998). Concordance (based on Z-scores provided by the Mantel test) declined as the interval between samples increased (Figure 6.6D), so that associations within a year were largely concordant, but associations across years within a season were concordant only in late summer. Overall, species associations were concordant for approximately half of the 18 intervals between snorkeling surveys. Associations were not typically changed by events like floods and droughts, but the second of two very severe droughts in three years coincided with distinct changes in associations of species or ecospecies. The empirical evidence of Matthews and Marsh-Matthews (2006a) suggests that although some smaller species groups (pairs, triads, or foursomes) might remain consistently in direct contact across years or even decades, there was little evidence that whole assemblage, multispecies associations were constant, or that selection pressures due to such multispecies groups would be consistent across such periods of time.

Persistence, Stability, and Control of Fish Assemblages

As pointed out at the beginning of the chapter, an important reason for trying to understand the levels of persistence and stability in fish assemblages is that assemblages that show high persistence, and particularly those with high stability, may be more influenced by biotic interactions (and have greater likelihood of deterministic control) than by abiotic factors (with greater likelihood of stochastic control). Clearly, the level of physical disturbance can influence the position of a community along a gradient of deterministic to stochastic control, as suggested by various authors, with biotic interactions likely more important in communities with low levels of disturbance and abiotic factors more important in communities with high levels of disturbance (e.g., Grossman et al. 1982; Peckarsky 1983). This is not to say that all assemblages showing stability are deterministically controlled or that all assemblages lacking stability are stochastically controlled. For instance, an assemblage with strong deterministic control based on competitive interactions among species could show a lack of stability because of differential time lags in the effects of species interactions (Strong 1983).

FIGURE 6.6. Persistence of associations among fish taxa in Brier Creek, Oklahoma, based on 19 snorkeling surveys taken over 22 years.

A. A simple example of similarity matrices based on two surveys with three taxa and three pools, where s_xy is the similarity of taxa x and y between two pools.

B. Actual data showing the strength of the associations between consecutive samples relative to the time between the surveys; the solid line is the regression line based on the array of Z-scores. Based on Matthews and Marsh-Matthews (2006a).

One way of addressing the question of the degree of deterministic control of fish assemblages would be to follow an assemblage over time in the absence of major disturbance. However, such systems are uncommon in nature, except for some isolated springs. An alternative approach would be to use a seminatural, artificial stream system. If all streams offer essentially the same environments, then deterministic control should result in high similarity among fish assemblages. Matthews and Marsh-Matthews (2006b) stocked seven outdoor artificial streams with identical numbers and kinds of species and then followed assemblage composition for 388 days. Even though the streams were as identical as possible, differences did develop over time in the extent of algal cover and in the level of predation. Different levels of predation were caused by the differential survival of sunfishes among pools. Somewhat surprisingly, even in the absence of natural disturbances, the assemblages diverged significantly in composition, so that the ultimate structure of any of the experimental assemblages “could not be predicted from its initial structure.” In other words, the study did not support predictions of strong deterministic control.

SUMMARY

Fish assemblages change over both ecological and evolutionary time scales. Assemblages controlled primarily by random processes (i.e., primacy of stochastic control) have greater variation in species composition and abundances, compared to those influenced primarily by nonrandom processes (i.e., primacy of deterministic control). Disturbances include any event that disrupts a community or a particular assemblage in some way. Measures of assemblage responses to disturbance include those of persistence (i.e., the presence or absence) of fish species or by stability, a measure that includes the kinds of species and their abundances. Responses of fishes to disturbance can be in the form of resistance to environmental stressors or through resilience—a measure of the ability of populations and assemblages to recover following the disturbance.

Studies of persistence and stability of fish assemblages are challenged because of greatly varying turnover rates in assemblages from different parts of North America. In addition, metrics used to determine stability form a hierarchical series of increasing sensitivity to change, such that the choice of analysis also influences the outcome. Furthermore, studies should be cognizant of biases. For example, studies on lotic systems are biased toward lowlatitude, southeastern regions, in contrast to those on lentic systems, which are biased toward higher latitudes.

SUPPLEMENTAL READING

Albanese, B. W., P. L. Angermeier, and J. T. Peterson. 2009. Does mobility explain variation in colonization and population recovery among stream fishes? Freshwater Biology 54:1444–60. Shows the variation in resilience, through recolonization ability, across streams and species.

Grossman, G. D., J. F. Dowd, and M. Crawford. 1990. Assemblage stability in stream fishes: a review. Environmental Management 14:661–71. A review of data on fish assemblage stability using coefficients of variation.

Grossman, G. D., P. B. Moyle, and J. O. Whitaker, Jr. 1982. Stochasticity in structural and functional characteristics of an Indiana stream fish assemblage: A test of community theory. The American Naturalist 120:423–54. An important paper that stimulated much research and discussion on the relative stability of fish assemblages.

Matthews, W. J., R. C. Cashner, and F. P. Gelwick. 1988. Stability and persistence of fish faunas and assemblages in three midwestern streams. Copeia 1988:945–55. Comparisons of fish assemblage persistence and stability using resemblance measures.

Matthews, W. J., and E. Marsh-Matthews. 2006. Temporal changes in replicated stream fish assemblages: Predictable or not? Freshwater Biology 51:1605–22. A test of stability of non-perturbed fish assemblages using an outdoor, experimental stream system.

Ecology of North American Freshwater Fishes

Подняться наверх