Grantee Research Project Results
Final Report: Assessment of Extinction Risk in Dynamic Landscapes
EPA Grant Number: R829090Title: Assessment of Extinction Risk in Dynamic Landscapes
Investigators: With, Kimberly A.
Institution: Kansas State University
EPA Project Officer: Hahn, Intaek
Project Period: October 29, 2001 through October 28, 2003
Project Amount: $219,415
RFA: Wildlife Risk Assessment (2001) RFA Text | Recipients Lists
Research Category: Ecological Indicators/Assessment/Restoration , Aquatic Ecosystems , Biology/Life Sciences
Objective:
The current global extinction crisis coincides with a period when landscapes are being transformed faster than ever before. Extinction risk needs to be assessed within a dynamic landscape context. The overall objective of this research project was to develop insights into how multiple stressors of landscape change (the rate, magnitude, and degree of habitat fragmentation) affected extinction risk for several generic neotropical migratory songbirds, a group of conservation concern. The specific objectives of this research project were to: (1) assess the extinction risk for different species-types in landscapes undergoing different rates of chronic habitat loss and fragmentation; (2) identify the vulnerability thresholds for species in dynamic landscapes; (3) determine the relative effects of different aspects of the disturbance regime on extinction risk; (4) develop screening criteria to provide a general assessment of extinction risk for species in dynamic landscapes; and (5) identify scenarios where adaptive responses to landscape change might occur and mitigate extinction risk. In addition to the original state objectives, we explored: (6) whether habitat restoration could recover or stabilize declining populations; and (7) how regional landscape connectivity (metalandscape connectivity) affected population viability and extinction risk in migratory songbirds.
Summary/Accomplishments (Outputs/Outcomes):
Assessment of Extinction Risk in Dynamic Landscapes
In the first phase of the project, we tackled the problem of assessing extinction risk for different types of migratory songbirds that varied in their sensitivity to habitat fragmentation by coupling a spatially structured avian demographic model with theoretical landscape models depicting different rates of habitat loss and fragmentation. Extinction risk was assessed relative to the vulnerability threshold, defined as the point where the change in population growth rate (Δλ) scaled to the rate of habitat loss (Δh) falls below –1 percent (Δλ/Δh = -0.01), analogous to the most conservative World Conservation Union criteria for assessing extinction risk. A species’ sensitivity to edge effects (the rate at which reproductive success declines in habitat fragments owing to increased nest predation and brood parasitism in small fragments) had the greatest effect on extinction risk, but species with intermediate edge sensitivity exhibited the greatest sensitivity to habitat fragmentation (greatest shift in vulnerability threshold). Counter to expectation, species appeared to persist across a greater range of habitat destruction when habitat loss was rapid (5% per year vs. 0.5% per year). This paradox results from a lagged population response to landscape change. With a 5 percent habitat loss/year, the landscape is changing more rapidly than the demographic potential of the species. Demographically, the population has already crossed the vulnerability threshold, but consists of the “living dead”; individuals persist, but there is insufficient habitat remaining to produce offspring to support a viable population. It takes up to three times longer to reach this threshold in landscapes subjected to 0.5 percent/year habitat loss, which would potentially provide more time for developing and implementing conservation strategies for reversing population declines. Because of the overriding importance of disturbance rate (landscape history) on species’ responses to landscape change, we discovered that the rate at which such thresholds are approached is dependent on how fast habitats were destroyed. The vulnerability threshold is both species- and landscape-specific, such that extinction risk for a given species on landscapes of similar structure may be very different depending on the history of landscape change (i.e., disturbance rate and length of disturbance). Temporal aspects of landscape change were more important than the spatial components of disturbance architecture for assessing extinction risk.
We have addressed the first four objectives of this project by: (1) assessing the effect of dynamic landscape change on extinction risk in neotropical migrants; (2) identifying vulnerability thresholds for various species in a dynamic landscape context; (3) determining the relative effects of different aspects of the landscape disturbance regime on extinction risk (the rate of disturbance had the greatest effect); and (4) developing simple screening criteria for assessing species’ extinction risk (edge-sensitivity is the most important predictor of extinction risk). This type of modeling approach can provide rapid assessments of species at risk. With the appropriate demographic data (species’ edge sensitivity and pertinent life-history parameters pertaining to clutch size, longevity, and survivorship) and information on the rate, magnitude, and duration of habitat destruction, this modeling approach could be implemented as an assessment tool to evaluate extinction risk for real species in real landscapes.
Demographic Limitations to Population Recovery Through Habitat Restoration
Habitat restoration is often recommended as a means of recovering declining populations without evaluating whether the species is habitat-limited and can be rescued by habitat restoration. We explored the magnitude of habitat restoration required to reverse population declines and recover populations to a viable state, which represents an application related to the initial problem (i.e., can we restore landscapes so as to stabilize or recover populations identified at risk of extinction?). Restoration efforts may be ineffective beyond some critical level of landscape disturbance and rate of population decline (restoration thresholds). Habitat restoration was simulated as a best-case scenario in which the landscape is instantaneously renewed to 100 percent habitat before, at, or after the species’ vulnerability threshold. The failure to stabilize or recover a population under this best-case scenario of habitat restoration would indicate that more modest levels and rates of restoration would not be sufficient for affecting a successful recovery. Habitat restoration was most effective for species with low to medium edge-sensitivities and in landscapes that had not previously experienced extensive fragmentation. For species in heavily fragmented landscapes, or species that are highly edge-sensitive, restoration needed to be initiated long before the vulnerability threshold (e.g., 20 years) to stabilize populations. Because of the erosion of demographic potential caused by habitat loss and fragmentation, populations cannot be recovered past this point through restoration alone (i.e., the restoration threshold), but would likely require additional conservation measures to increase survivorship, fecundity, or immigration rates among populations. Past the restoration threshold, population recovery is limited by demographic factors rather than the amount of habitat. In practice, however, habitat restoration is generally not initiated until a population is at risk of extinction, but our model results demonstrate that restoration needs to be done proactively and with consideration of demographic factors other than sensitivity to habitat loss if it is to be successful in cases where populations are demographically limited rather than habitat limited. Evidence that populations decline in response to habitat loss is not sufficient to conclude that habitat restoration will recover the population.
The Implications of Metalandscape Connectivity For Population Viability
The continued persistence of many songbirds in the face of widespread habitat loss and fragmentation suggests that regional landscape connectivity is critically important for the local occurrence of these species. Many, perhaps most, migratory songbirds are believed to exist as coupled source-sink populations, where immigration from source landscapes is responsible for maintaining populations that are no longer viable (sink populations). Extinction risk is mitigated for these species at a broader regional scale (if not at a local landscape scale) where immigration rates are sufficient to offset local population declines resulting from habitat loss and fragmentation. For declining populations that are sensitive to fragmentation effects (high edge-sensitivity), immigration rates (≥ 40%) are required to maintain viable populations once more than 30 percent of the habitat has been destroyed on the landscape. Although high, this is within the range of immigration rates that have been reported for real populations of migratory birds, suggesting that high metalandscape connectivity may indeed contribute to the persistence of these species at a landscape scale despite ongoing habitat loss and fragmentation in a given landscape.
Adaptive Responses to Landscape Change
Adaptive response to landscape change is probably unlikely in species that exhibit regional source-sink population dynamics, such as migratory songbirds. Any adaptations that emerge in sink (fragmented) landscapes that counteract the negative effects of fragmentation, such as increased brood parasitism, will be swamped by immigrants from outlying source (unfragmented) landscapes where selection pressures are less stringent. Our model results demonstrated that even in closed (isolated) populations, adaptive responses (e.g., a shift from high to low edge-sensitivity) were unlikely to fully recover declining populations even when such responses occurred rapidly, within three to five generations. Thus, adaptive responses to landscape change are unlikely to occur, or are unlikely to occur quickly enough, to mitigate extinction risk for neotropical migratory songbirds in landscapes subjected to habitat loss and fragmentation.
Significance of Findings
Assessments of extinction risk typically ignore landscape history. Instead, species risk assessments are based either on the capacity of the current landscape to support a viable population or on the probability that the population will persist under various scenarios of future land-use change or proposed land-management practices. Assessments of extinction risk based on current landscape patterns may not be accurate without information on the history of landscape change. A species’ risk of extinction on landscapes that appear similar structurally might be very different depending on the rate of disturbance and the amount of time the landscape took to reach that state. Thus, traditional landscape metrics based on the current amount and fragmentation of habitat may not be sufficient for evaluating extinction risk without invoking the “ghost of landscapes past.” Identification of “vulnerability thresholds,” where the change in population growth rate (Δλ) scaled to the rate of habitat loss (Δh) falls below –1 percent (Δλ/Δh = -0.01), provides a more conservative measure of extinction risk than the extinction (persistence) threshold and would thus permit action to be taken sooner, resulting in a proactive rather than reactive response to the conservation of declining populations.
Habitat restoration often is recommended as a means of conserving declining species. Rigorous assessments, however, are seldom made of the potential for habitat restoration to recover target populations. Our findings suggest that for species that are very sensitive to habitat fragmentation, habitat restoration would need to be initiated well before the population is assessed as being at risk of extinction (i.e., before the vulnerability threshold is crossed). Thus, by the time a species is identified as being at risk of extinction, it may already be too late. In this case, such species are demographically limited rather than habitat limited and, therefore, habitat restoration cannot restore the population to a viable state. Other conservation measures would be needed to boost reproductive output or immigration rates and survivorship.
For highly mobile animal populations, such as migratory songbirds, high rates of immigration help to maintain local populations despite significant habitat loss and fragmentation within those landscapes. Source-sink landscape dynamics are critically important to population viability at a regional scale. It is to be expected, however, that widespread habitat loss throughout a region will eventually compromise this “rescue effect” as more landscapes are converted to population sinks. It is not realistic to assume that species will be able to evolve or develop adaptive responses to landscape change that will permit them to recover. Efforts must therefore be directed toward conserving or restoring sufficient amounts of habitat (well above a species’ vulnerability threshold) and slowing the rate of habitat destruction at a broad enough scale to maintain the metalandscape connectivity and regional source-sink dynamics of neotropical migratory songbirds.
Journal Articles on this Report : 2 Displayed | Download in RIS Format
Other project views: | All 13 publications | 7 publications in selected types | All 3 journal articles |
---|
Type | Citation | ||
---|---|---|---|
|
Schrott GR, With KA, King AW. On the importance of landscape history for assessing extinction risk. Ecological Applications 2005;15(2):493-506. |
R829090 (Final) |
Exit Exit |
|
Schrott GR, With KA, King AW. Demographic limitations of the ability of habitat restoration to rescue declining populations. Conservation Biology 2005;19(4):1181-1193. |
R829090 (Final) |
Exit Exit |
Supplemental Keywords:
risk assessment, ecological effects, animal, population, terrestrial, ecology, modeling, global extinction, dynamic landscape context, landscape change, habit fragmentation, vulnerability thresholds, extinction risk, habitat restoration,, RFA, Scientific Discipline, Ecosystem Protection/Environmental Exposure & Risk, Ecosystem/Assessment/Indicators, Ecosystem Protection, wildlife, Ecological Effects - Environmental Exposure & Risk, Environmental Monitoring, Habitat, Ecological Risk Assessment, Ecology and Ecosystems, ecological exposure, habitat dynamics, landscapes, extinction risk, dynamic landscapes, assessment models, adverse impacts, stressors, environmental stressor, extinction risk in dynamic landscapes, habitat disturbance, habitat loss, invasive species, assessment methodsRelevant Websites:
http://www.ksu.edu/withlab Exit
Progress and Final Reports:
Original AbstractThe perspectives, information and conclusions conveyed in research project abstracts, progress reports, final reports, journal abstracts and journal publications convey the viewpoints of the principal investigator and may not represent the views and policies of ORD and EPA. Conclusions drawn by the principal investigators have not been reviewed by the Agency.