disturbance

Edward McCorkindale, MWA and MPA Candidate www.unm.edu/~emccorkindale Professional Profile e: emccorkindale@unm.edu
 * Disturbance-Adapted Ecosystems**

Riparian ecosystems are adapted to multiple disturbance types. Plant community succession in riparian areas of the southwestern United States are characterized by disturbance, which have changed due to anthropogenic changes to natural regimes of flood, geomorphic, and fire disturbance regimes. Restoration in these systems should have a focus on restoring, or at the very least, mimicking natural disturbance. At the same time, restoration prescriptions must not be made before there is a clear understanding of how the ecosystem will respond to the introduction or removal of a disturbance. Restoration activities should also incorporate an understanding of upstream natural and anthropogenic disturbances. Even in cases where natural disturbance mechanisms do not change due to human activities, disconnection between upland/upstream/forest and downstream/riparian areas (e.g. levees, forest buffers, dams) can change downstream ecological processes.
 * //Abstract //**

//Introduction //
Riparian ecosystems are adapted to multiple disturbance types, including fire, floods, deposition, and erosion. Organisms that are endemic to riparian ecosystems have evolved within natural riparian disturbance regimes, with life histories that incorporate or even depend on these disturbances. Every disturbance type is different, although the impact of two different disturbance events may be similar based on their duration, their scope, and the natural biotic and abiotic processes that they affect. A traditional definition of disturbance has focused on irregular, usually catastrophic, events which cause abrupt changes in the structure and processes of natural communities (White, 1979). White points out two limitations with this definition: 1) Some kinds of disturbances are initiated or carried out by natural components of the community; and 2) There is a gradient between minor and major changes in natural communities. First, one must differentiate between natural and anthropogenic disturbances. Unless otherwise noted, “disturbance” normally refers to changes that are initiated and carried out by natural processes (e.g. flooding due to significant rainfall). Second, Sousa (1984) also notes that changes in a community can vary in their magnitude, from negligible to catastrophic or extreme. Therefore, it can be difficult to determine whether changes are a simple ecological process in the community, (e.g. physiological acclimatization) or if they are indeed a disturbance. One must look at changes in the community as disturbances on multiple size (negligible to catastrophic), spatial (microscopic to regional), and temporal (hourly to millennial) scales. Sousa also notes that the traditional definition often suggests that natural communities persist in equilibrium conditions between disturbances. Instead, evidence suggests that it is difficult to differentiate between communities that are in equilibrium and those that are not. Therefore, one must understand that disturbances are ongoing, that natural disturbances happen on different temporal, spatial, and size scales, and that few communities have an equilibrium or static state.

//Types of natural disturbance // Natural disturbances can be classified in broad categories based on the mechanism of disturbance (e.g. Rood, et al., 2007: fire, water/flood, ice), differentiated by whether they are biotic and abiotic disturbances (Swanson, 1994), or listed exhaustively as individual types (White, 1979). In some cases, the scope of study may be focused only on the types of disturbance that affect a particular ecosystem or organism (Scott et al, 1997), or the type of disturbance may not be as important as the impacts of all disturbances in general (Sousa, 1984). Table 1 is a list of natural disturbances compiled from multiple sources. Disturbance types noted with an asterisk are those that are particularly important for riparian ecosystems, and that will be investigated in greater detail in this article. Ultimately, one must recognize that multiple natural disturbances are at work in every system, that different disturbance types interact to uniquely define certain ecosystems, and that the system responses to disturbance triggers can mimic each other even though the disturbance mechanisms may be very different.


 * Table 1.** //Disturbance types can mimic each other even though the disturbance mechanisms may be very different.//

The ability to effectively maintain and restore river ecosystems is seated on a scientific understanding of how ecosystems function and change, and the impact of anthropogenic activities on ecosystem functions. Both natural and human disturbances are observed on spectrums of size, spatial, and temporal scales. Combinations of both smaller scale upland terrestrial and riverine disturbances move river ecosystems towards greater complexity, higher biodiversity, and will increase the probability that the system will recover from catastrophic (larger scale) disturbances (Swanson, 1994). Sousa (1984) states that disturbance regimes are characterized by the forces and responses being studied, and most notably using: areal extent; magnitude; frequency; predictability; turnover rate; or rotation period. Sousa also notes that regimes are connected to patterns of succession of organisms in a system, and that the heterogeneity in structure and processes in a system are due to the interaction of different disturbance regimes.

//Disturbance-adapted ecosystems, and riparian disturbance regimes in the southwestern United States //
In the vein of a traditional view of disturbance, Williams et al (1998) depicts disturbance, along with climate, site hydrology, and plant function types, as factors that interact with and control the traits of riparian ecosystems. This view puts natural and human-related disturbances under the same heading, as external perturbations that have the ability to degrade, or promote, processes controlling the stability of riparian ecosystems. Dwire & Kauffman (2003) similarly identifies hydrology, climate, disturbance, vegetation, and geomorphology as components, but each component is connected to the others through interactions that contribute to the structure, function, and composition of riparian ecosystems. As noted in the definitions of disturbance by Sousa (1984) and the natural flow regime by Poff et al (1997), riparian systems should be looked at as dynamic systems with no equilibrium state. Further, since climate, site hydrology, biotic functions, and disturbances interact with each other within riparian ecosystems, and in some cases are one and the same (e.g. flooding is both a type of disturbance, and a characteristic of site hydrology), riparian areas must be looked at as disturbance-adapted ecosystems, rather than being "disturbance-affected."

Anderson (1992) (as cited in Swanson, 1994) notes that aquatic systems often experience quick recovery from natural disturbances because they are usually scattered in spatial extent, leaving many undisturbed reaches of the river that can serve as refuges for aquatic and riparian organisms that will eventually reoccupy disturbed areas. Abiotic elements, such as the mobilization and deposition of sediment, and flow of water, may be looked at this way as well: the instability of where water and sediment are located over time maintains the heterogeneity of riparian ecosystems so that there are continually new areas for colonization (e.g. bare sediment, temporarily inundated channels) as well as established areas of more mature vegetation and habitat.

Plant community successional processes are characterized, in part, by disturbance (Halofsky & Hibbs, 2009). In floodplains, where flooding is a prominent disturbance mechanism, tolerance to flooding may be a major factor influencing tree distribution (Hughes, 1990). In arid and semi-arid areas of North America, like the southwestern U.S., newly formed surfaces, such as scoured riparian areas and flood deposits, become dominated by early successional woody plants, particularly cottonwoods (//Populus// spp.) and willows (//Salix// spp.) (Scott, et al, 1997; Hughes, 1990). Rood et al (2007) observes that riparian cottonwoods are tolerant of, and dependent on, physical disturbance for population rejuvenation. Further, variability in disturbance responses (in the form of different disturbance types and scales) creates niche differentiation between species. Therefore, diverse disturbances can create heterogeneity in patterns of cottonwood establishment. For example, in the middle portion of the Rio Grande in New Mexico, river/riparian dynamics have been identified as a factor in the ability of native vegetation to establish and survive in riparian areas (MRGESCP, 2004). In the Middle Rio Grande, changes in the frequency, magnitude, duration, and timing of flows have impacted the life histories of several cottonwood and willow species, as well as exotic species, such as Russian olive and tamarisk (Figure 1). Native species are more dependent on well-timed flows than exotic species, since the latter can colonize under a wider range of conditions throughout the year. On the other hand, natural disturbance regimes likely favor native cottonwoods and disfavor some invasive woody plants (Rood et al, 2007). Palmer et al (2005) observes that scouring floods can enhance biodiversity by reducing the abundance of more competitive species that favor stable flows.


 * Figure 1.** //Periods of seed dispersal and seed viability span different parts of the year for plant species in the Rio Grande, New Mexico. Changes in the frequency, magnitude, duration, and timing of flows have impacted the life histories of cottonwood and willow species, as well as exotic species, such as Russian olive, Siberian elm, and Saltcedar. Adapted from MRGESCP (2004), Figure 3-10.//

Disturbances serve multiple ecological roles. For riparian ecosystems, floods, fire, ice, and geomorphic changes are the primary disturbance mechanisms/types that occur in-stream and in riparian and upland areas (Table 2). Swanson (1994) notes that stream channel changes are the dominant disturbance process in riparian zones. Geomorphic disturbances (such as bank collapse, and low-flow deposition) in unconstrained reaches can be reflected by varied composition of vegetation, meandering of the stream channel (lateral changes), and where the gradient or velocity decreases, deposition of sediment and debris in alluvial fans, islands, bars, and the stream bed. For example, bank collapses and landslides can deflect the course of the river to the opposite side (particularly in narrower valleys) so that the river pinches the opposite bank, potentially causing erosion of the opposite bank and an increase in sinuosity of the channel (Swanson, 1994).

**Table 2.** //In riparian ecosystems, floods, geomorphic changes, fire, and ice are the primary disturbance mechanisms. These processes have great ecological importance for these systems. Anthropogenic activities have altered many of these mechanisms, although scientific understanding of disturbance has provided concepts that can be used in restoring riparian areas and their active and historic disturbance regimes.//

Halofsky & Hibbs (2009) and Dwire & Kauffman (2003) note that riparian plant communities are resilient to fire disturbance and little post-fire rehabilitation is necessary for continued function, unless there has already been extensive degradation to the ecosystem. Ellis (2001) (as cited in Dwire & Kauffman, 2003) describes a study of cottonwoods (//Populus deltoids wislizenii//) and willows (//Salix gooddingii//) in New Mexico that focuses on root suckering as a method for these species to regenerate after fire disturbance. Over 40% of cottonwoods produced shoots that survived at least 2 years following fire, and about 73% of willows produced shoots within 4 months. Considering that only 55% of exotic tamarisks (//Tamarix ramosissima//) produced sprouts, this may indicate differential responses to fire disturbance that favor native species over exotics. Willows and cottonwoods can also become established in high densities in burned riparian areas via anemochory (wind) or hydrochory (fluvial delivery) (Dwire & Kauffman, 2003). Flood and fire regimes interact where they happen close in time, such that riparian plant succession can be changed by greater flood frequency in post-fire systems (Halofsky & Hibbs, 2009).

Rood et al (2007) notes that ice drives, following spring break-up of frozen rivers, can be the most powerful of geomorphic agents acting on a system. Therefore, extensive scouring, shearing, and deposition can take place. This type of disturbance can only be expected to occur regularly on northern and north-flowing rivers, where a late spring thaw of river ice allows a sequence of ice break-up, downstream passage, and then jamming as the ice encounters more secure ice. These jams can act as temporary dams which, when the pressure upstream reaches the dams breaking point, release surges of ice and water (Rood et al, 2007). This sequence can cause localized scouring and overbank flooding, which usually occurs prior to increasing stream flows. Therefore, this processes has heterogeneity in its affects on different reaches of the river.

Anthropogenic disruptions to natural hydrology in rivers can change the dynamic equilibrium between geomorphic and hydrologic ecological processes that are listed in Table 2. In systems with natural flood retention, modification of this function can instead route flows quickly downstream, thereby increasing the magnitude and frequency of floods but reducing baseflows (Poff et al, 1997). This process intersects with the ecological of native species, as the duration of higher flows can be reduced by removal of natural flood retention processes. Similar to fire disturbance (Dwire & Kauffman, 2003), biological and botanical tolerance to prolonged flooding allows many native species to persist in locations where they might otherwise be outcompeted by less tolerant, and possibly exotic, species. Further, the natural timing of flows provides cues for life cycle transitions in fish.

//Disturbance and habitat restoration //
Decreasing interactions between river, riparian, and other terrestrial processes can result in a decline in the biotic potential of the system (Kauffman, et al, 1995). As scientific understanding of a system is developed, a regime of disturbance can be distilled that can be used during river restoration activities such that management of the system may emulate natural disturbance. Active river restoration actions should not take place without a clear understanding of how the ecosystem will respond to the removal of disturbance factors (Kauffman et al, 1995). Passive restoration may capitalize on the resilience of riparian ecosystems by allowing them to recover after anthropogenic disturbances have been removed. Palmer et al (2005) proposed that river restoration projects, if they are to be ecologically sound, must be based on a goal of a more dynamic river that is self-sustaining and resilient to disturbance. That is, projects must involve restoration of natural river processes, including natural disturbance regimes (Table 3).

//Adapted from "Ecologically sound restoration: avoiding ineffective approaches," in Palmer et al (2005), 212-213.//
 * Table 3.** //Ineffective approaches to riparian habitat restoration include techniques that fail to incorporate disturbance.//

The patchy nature of natural disturbance creates varying conditions within overall ecosystems, where areas for colonization, habitat, and succession create heterogenous systems (Kreutzweiser et al, 2012). Poff et al (1997) references a number of recent river restoration efforts (within the last 20 years) that have sought to establish more natural flow regimes, where there is variability in flows and geomorphic processes that are critical to ecological functions. Where full restoration isn't possible, these processes can be "mimicked" by managing the magnitude, frequency, duration, timing, and extent of disturbances. For example, the mimicking of timing and duration of floodplain inundation can be accomplished by managed flow pulses and irrigation, thereby stimulating recruitment of native riparian plants, nutrient deposition, and life cycle transition cues (Molles et al, 1995, Robertson, 1997, as cited in Poff et al, 1997). Rood et al (2007) notes that flooding in upland forests is similar to fire disturbance, because both disturbance mechanisms result in dramatic short-term changes, but also enable subsequent ecosystem rejuvenation. When emulating fire disturbance, Nitschke (2005) indicates that, ultimately, the goals of restoration are essential for determining how to use riparian forest harvesting as the emulating practice. This study observes that different harvesting practices (e.g. clear-cutting, selective cutting) emulate fire in different ways. For example, while clear-cutting is emulates the effect that wildfire has on water temperature, increased summer flows followed by a return to previous conditions are better emulated by partial harvesting. On the other hand, water chemistry changes due to fire are not easily emulated by harvesting. Nitschke suggests that harvesting be combined with other practices, such as riparian buffers, to better emulate fire. This study cautions that ignoring water chemistry changes in management decisions can be very serious due to long-term impacts of these changes.

//Mimicking flood disturbance in the Middle Rio Grande, New Mexico //
Chistensen et al (1996) (as cited by Molles et al, 1998) labels all riparian forests in the United States as threatened ecosystems, and riparian forests in New Mexico as endangered. Molles (1998) observes that the extensive cottonwood-willow forest along the middle Rio Grande in central New Mexico is largely a legacy of a historical flooding regime. Due to anthropogenic flood control operations, new stands of cottonwoods are not becoming naturally established, and stabilized flows are favoring non-native tree species. Two approaches to river restoration are identified by Molles: first, conversion of non-native forests to native forest (habitat restoration); and second, reconnecting disconnected areas of the floodplain through managed flooding (mimicking the natural flow regime). In this study, managed floods are used to mimic conditions of a lower-energy flood in the backwater of a larger flood. While the study focuses on ecological responses on the forest floor, it provides some overarching observations regarding using the latter restoration method to reestablish floodplain connections.

Molles et al (1998) uses the concept of ecosystem "reorganization" through which disconnected riparian areas can be rejuvenated by managed floods in order to reach a "steady-state phase" of approximate conditions prior to flood control. This article notes that "steady-state" refers to reestablishing ecological processes found under a natural flood disturbance regime, and not a static system. The study finds that the reorganization period following managed floods are different for each ecosystem component. Therefore, some ecosystem functions, once being (re)subjected to the natural flow regime, may take time before they return to the way that they were before flood control. The most dramatic responses to reconnection are observed in microbial populations and unseen chemical and physical processes. While this study does not address vegetation responses to river reconnection, it indicates that smaller scale return of ecological processes under a natural flow regime may provide a foundation for larger scale restoration that, while taking longer to occur, is resilient and self-sustaining.

//Conclusions and suggestions //
Kauffman et al (1995) notes that previous restoration engineering has focused on in-channel modifications, while mostly ignoring interactions between terrestrial (riparian) and riverine processes. A robust scientific understanding of an ecosystem, and its natural disturbance regimes, is necessary for its restoration. Restoration actions that do not address continuing limitations to recovery (e.g. removal of the wrong anthropogenic modifications, rather ones that are affecting the system the most) can lead to a misinterpretation of ecosystem needs, resulting in manipulations that continue to degrade the system. Additionally, a scientific understanding of an ecosystem should include identifying its "resilience threshold" (Nitschke, 2005), which may allow informed tradeoffs between management that emulates natural disturbance and allowing natural disturbance to drive management (passive restoration).

The ability to define disturbance regimes and the effects of disturbance, particularly wildfire (Nitschke, 2005) is limited by the stochastic nature of these processes. Sibley et al (2012) asks whether restoration that emulates natural disturbance should have a goal of reflecting the true variability of disturbances or the long-term average of magnitude, frequency, timing, and extent of disturbances. Although there are scientific limits to how specific a natural flow regime can be defined and restored, Poff et al (1997) cautions against managing for average conditions because these conditions may be based on high variability that are lost by using the average. Poff also notes that this is particularly important for systems where geomorphic and ecological processes have a nonlinear response to river flows. For example, banks only overflow above a certain river discharge, and not only at half of the peak discharge. Another caveat for emulating disturbance (Nitschke, 2005) is that there needs to be a balance between emulation and the social acceptability of the disturbance mechanism that is being used. In particular, wildfire, and even controlled fires, may be unacceptable as a practice, so other practices, however less-congruent that they are with the disturbance being emulated, may be the preferred option.

Ultimately, restoration practitioners have a need to understand the interconnections between natural and anthropogenic disturbances, and their roles among other ecosystem processes. Kreutzweiser et al (2012) notes that there are connections between upland/upstream ecosystems and their disturbances, and downstream riparian disturbance and renewal processes. Forest disturbances promote aquatic habitat heterogeneity and biological diversity. Therefore, this article observes that restoration activities should incorporate an understanding of upstream natural and anthropogenic disturbances. Even in cases where disturbance mechanisms do not change due to human activities, disconnection between upland/upstream/forest and downstream/riparian areas (e.g. levees, forest buffers, dams) can change downstream ecological processes.

//References //
> //Conservation//, Issue 97, pp. 159-170. > Princeton Scientific Publishing Co, pp. 119-152.
 * Chien, N., 1985. Changes in river regine after the construction of upstream reservoirs. //Earth Surface Processes and Landforms//, Issue 10, pp. 143-159.
 * Dwire, K. A. & Kauffman, J. B., 2003. Fire and Riparian Ecosystems in Landscapes of the Western USA. Forest Ecology and Management, Volume 178, pp. 61-74.
 * Ellis, L., 2001. Short-term response of woody plants to fire in a Rio Grande riparian forest, Central New Mexico. //Biological//
 * Halofsky, Jessica E., and David E. Hobbs, 2009. Controls on Early Post-fire Woody Plan Colonization in Riparian Areas. //Forest Ecology and Management// 258: 1350-1358.
 * Hughes, F. M. R., 1990. The Influence of Flooding Regimes on Forest Distribution and Composition in the Tana River Floodplain, Kenya. //Journal of Applied Ecology//, Issue 27, pp. 475-491.
 * Johnson, W. C., 1994. Woodland expanson in the Platte River, Nebraska: patterns and causes. //Ecological Monographs//, Issue 64, pp. 45-84.
 * Kauffman, J. B. et al, 1995. Ecological Approaches to Riparian Restoration in Northeast Oregon. Restoration & Management Notes, Summer.13(1).
 * Kondolf, G. M. & Curry, R. R., 1986. Channel erosion along the Carmel River, Monterey County, California. //Earth Surface Processes and Landforms//, Issue 11, pp. 307-319.
 * Kreutzweiser, David P., Paul K. Sibley, John S. Richardson, and Andrew M. Gordon, 2012. Introduction and a Theoretical Basis for Using Disturbance by Forest Management Activities to Sustain Aquatic Ecosystems. //Freshwater Science// 31(1):224-231.
 * Molles, M. C. et al., 1998. Managed Flooding for Riparian Ecosystem Restoration. //BioScience//, 48(9), pp. 749-756.
 * Molles, M. C., Crawford, C. S. & Ellis, L. M., 1995. Effects of an experimental flood on litter dynamics in the Middle Rio Grande riparian ecosystem. //Regulated Rivers: Research & Management//, Issue 11, pp. 275-281.
 * <span style="font-family: Tahoma,Geneva,sans-serif;">MRGESCP, 2004. Habitat Restoration Plan for the Middle Rio Grande. September.
 * Nitschke, Craig R., 2005. Does Forest Harvesting Emulate Fire Disturbance? A Comparison of Effects on Selected Attributes in Coniferous-Dominated Headwater Systems. //Forest Ecology and Management// 214:305-319.
 * <span style="font-family: Tahoma,Geneva,sans-serif;">Palmer, M. et al, 2005. Standards for Ecologically Successful River Restoration. //Journal of Applied Ecology//, Issue 42, pp. 208-217.
 * <span style="font-family: Tahoma,Geneva,sans-serif;">Prestegaard, K. L., 1988. Morphological controls on sediment delivery pathways. In: D. E. Walling, ed. //Sediment budgets//. International Association of Hydrological Sciences Publication nr 174 ed. Wallingford (UK): IAHS Press, pp. 533-540.
 * Rood, S. B. et al, 2007. Floods, fire, and ice: disturbance ecology of riparian cottonwoods. //Canadian Journal of Biology//, Issue 85, pp. 1019-1032.
 * <span style="font-family: Tahoma,Geneva,sans-serif;">Roberston, L., 1997. //Water Operations on the Pecos River, New Mexico, and the Pecos bluntnose shiner, a federally-listed minnow//. US Conference on Irrigation and Drainage Symposium.
 * Scott, M. L., Augle, G. T. & Friedman, J. M., 1997. Flood Dependency of Cottonwood Establishment Along the Missouri River, Montana, USA. Ecological Applications, May, 7(2), pp. 677-690.
 * Sibley, Paul K., David P. Kreutzweiser, Brian J. Naylor, John S. Richardson, and Andrew M. Gordon, 2012. Emulation of Natural Disturbance (END) for Riparian Forest Management: Synthesis and Recommendations. //Freshwater Science// 31(1):258-264.
 * Sousa, W. P., 1984. The Role of Disturbance in Natural Communities. Annual Review of Ecology and Systematics, Volume 15, pp. 353-391.
 * <span style="font-family: Tahoma,Geneva,sans-serif;">Sparks, R. E., 1992. Risks of altering the hydrologic regime of large rivers. In: J. Cairns, B. R. Niederlehner & D. R. Orvos, eds. //Predicting ecosystem risk//. //Advances in modern environmental toxicit//y. Princeton (NJ):
 * Swanson, F., 1994. Natural Disturbance Effects on Riparian Areas. Natural Resources and Environmental Issues, 1(3).
 * White, P. S., 1979. Pattern, Process, and Natural Disturbance in Vegetation. Botanical Review, Jul. - Sep., 45(3), pp. 229-299.
 * <span style="font-family: Tahoma,Geneva,sans-serif;">Williams, D. G., Brunel, J. P., Schaeffer, S. M. & Snyder, K. A., 11-16 Jan 1998. //Biotic Controls Over Function of Desert Riparian Ecosystems//. Phoenix, Arizona, U.S. Department of Agriculture.
 * Williams, G. P. & Wolman, M. G., 1984. //Downstream effects of dams on alluvial rivers,// Reston (VA): US Geological Survey.