globalchange  > 气候变化事实与影响
DOI: doi:10.1038/nclimate2252
论文题名:
Invasive hybridization in a threatened species is accelerated by climate change
作者: Clint C. Muhlfeld
刊名: Nature Climate Change
ISSN: 1758-1305X
EISSN: 1758-7425
出版年: 2014-05-25
卷: Volume:4, 页码:Pages:620;624 (2014)
语种: 英语
英文关键词: Ecological genetics ; Evolutionary ecology ; Climate-change ecology ; Invasive species
英文摘要:

Climate change will decrease worldwide biodiversity through a number of potential pathways1, including invasive hybridization2 (cross-breeding between invasive and native species). How climate warming influences the spread of hybridization and loss of native genomes poses difficult ecological and evolutionary questions with little empirical information to guide conservation management decisions3. Here we combine long-term genetic monitoring data with high-resolution climate and stream temperature predictions to evaluate how recent climate warming has influenced the spatio-temporal spread of human-mediated hybridization between threatened native westslope cutthroat trout (Oncorhynchus clarkii lewisi) and non-native rainbow trout (Oncorhynchus mykiss), the world’s most widely introduced invasive fish4. Despite widespread release of millions of rainbow trout over the past century within the Flathead River system5, a large relatively pristine watershed in western North America, historical samples revealed that hybridization was prevalent only in one (source) population. During a subsequent 30-year period of accelerated warming, hybridization spread rapidly and was strongly linked to interactions between climatic drivers—precipitation and temperature—and distance to the source population. Specifically, decreases in spring precipitation and increases in summer stream temperature probably promoted upstream expansion of hybridization throughout the system. This study shows that rapid climate warming can exacerbate interactions between native and non-native species through invasive hybridization, which could spell genomic extinction for many species.

Changes in species ecology associated with climate change have been documented for a broad range of organisms1, 6, yet empirical understanding of how climate change influences evolutionary processes and resulting patterns of biodiversity is limited. One consequence of climate-induced range shifts is increased sympatry between previously isolated species, potentially resulting in introgressive hybridization (genes from an invasive species spread into a native species)7, 8. Climate-induced expansions of introgression have been predicted for many terrestrial and aquatic species, especially species that are sensitive to temperature and streamflow conditions2, 6, 9. Although hybridization can increase the adaptive potential of closely related species through periods of climate change7, hybridization driven by human activities, such as translocation of species, tends to occur quickly and reduce fitness10, genomic integrity11, and ultimately native species diversity12. Despite predictions that interspecific hybridization may increase as a result of species range shifts and human impacts, empirical evidence linking such evolutionary changes to recent climatic change is extremely scarce3, 13.

Salmonids—a group of fishes of enormous ecological and socio-economic value—are ideal organisms for examining how climate change facilitates hybridization between native and non-native species. Hybridization and introgression are particularly common between salmonids and other fish because there are limited pre- or post-zygotic barriers to introgression12, and widespread introductions have created sympatry between many previously allopatric species11. Moreover, the distribution, abundance and phenology of salmonid fishes are strongly influenced by climatic conditions through species-specific adaptations to water temperature and the timing and magnitude of streamflow14. Thus, ongoing climate change is expected to differentially affect salmonid species, possibly expanding zones of introgressive hybridization as some species expand their distribution and increase in abundance during periods of warming and shifting hydrologic regimes14, 15.

Quantifying spatial and temporal genetic changes in wild populations, including introgression, provides strong support for climate-induced evolutionary change13, 16. Such data, however, are limited among vertebrates, especially for rare and endangered species. Here we use long-term genetic monitoring data (1978–2008) to test the prediction that climatic variation has affected the spread of introgressive hybridization between threatened native westslope cutthroat trout and non-native rainbow trout. Cutthroat trout (Oncorhynchus clarkii) and introduced rainbow trout can overlap in time and space during spring-spawning and produce fertile offspring when they interbreed17. Introgression often continues until a hybrid swarm (a randomly mating population containing only hybrid individuals) is formed and all the native genomes are lost18. Introgression poses a serious threat to all subspecies of inland cutthroat trout in western North America as a result of widespread stocking of rainbow trout into historical cutthroat trout habitats; two subspecies are now extinct and five are listed as threatened under the US Endangered Species Act. The westslope cutthroat trout is the most widely distributed subspecies, and hybridization is the leading threat to the persistence of genetically pure populations; known non-hybridized populations occupy less than 10% of their historic range19.

Relative to cutthroat trout, rainbow trout prefer warmer temperatures, lower spring flows, earlier spring runoff, and tolerate greater environmental disturbance15, 17, 20, 21. Therefore, we tested the prediction that increased summer stream temperatures, decreased spring precipitation, and wildfire disturbance have influenced the spatiotemporal spread of hybridization throughout the Flathead River system (USA and Canada; Supplementary Fig. 1 and Supplementary Table 1). The drainage presents an ideal location to examine this prediction because it is one of America’s most pristine river systems22 and does not contain other introduced species that may influence interactions between these species. However, the basin has warmed considerably over the past century, with annual average temperatures warming at two times the global average23. From 1948 to 2008, mean annual air temperatures increased by 0.82 °C, a rate of 0.14 °C/decade, yet from 1978 to 2008 warming nearly tripled to 0.36 °C/decade24. Increasing air temperatures are contributing to a decrease in spring snowpack and a shift towards an earlier spring season, resulting in peak spring runoff two to three weeks earlier than the historic average and lower spring and summer flows23, 24, 25, which are directly contributing to increased summer stream temperatures26, 27. Changes in spring temperature and precipitation correspond strongly to these observed changes and significantly influence the timing and magnitude of streamflow in the basin25, 28. Furthermore, wildfire has burned approximately 16% of the basin from 1984 to 2008, potentially increasing the rate of stream warming by removing canopy cover24. Such changes are hypothesized to benefit rainbow trout, as spawning and recruitment are limited by high spring flows and cold water temperatures15.

Over 20 million rainbow trout were stocked throughout the Flathead River system, primarily in low elevation areas, beginning in the late 1800s and ending in 19695. Despite massive stocking efforts, genetic samples collected from the late 1970s and early 1980s detected low levels (<2%) of hybridization in just 2 of 20 sites, and a hybrid swarm (representing multiple generations of hybridization) with a predominant (92%) genetic contribution from rainbow trout was discovered in the lower valley in 1994 (Fig. 1a and Supplementary Table 1)5, 29. Conversely, samples from the 2000s showed introgression in 9 of the 18 previously non-hybridized sites, demonstrating that introgression increased rapidly over a 30-year period (from 10% to 52% of all sites; Fig. 1b). In the most extreme example, introgression in one population increased tenfold from 3% to 33%.

Figure 1: Spatiotemporal spread of hybridization relative to climatic changes.
Spatiotemporal spread of hybridization relative to climatic changes.

ad, Maps showing the spread of rainbow trout hybridization in relation to average decadal summer stream temperature (a,b) and May precipitation (c,d); 1980s (a,c) and 2000s (b,d). Sample data are included in Supplementary Table 1.

Historical (1978–1984) samples were genotyped at six species diagnostic allozyme loci5, 29. Recent (2000–2008) samples from the 21 historically sampled populations and a further 29 populations throughout the basin were genotyped at seven diagnostic microsatellite loci10, 30 (Supplementary Table 1). A species diagnostic locus has non-overlapping allele sizes in the two parental taxa. For our purposes, individual trout could have zero, one or two rainbow trout alleles at each locus. These molecular markers have identical statistical properties for estimating introgression at the population level (that is, detection probability does not vary by marker type). However, by using more diagnostic loci in recent samples the sampling error around our estimates is slightly reduced and there is a small increase in power to detect low levels of hybridization. With a sample size of 25 individuals (average sample size in this dataset) the power to detect 1% rainbow trout admixture is 0.95 with six diagnostic loci, 0.97 with seven diagnostic loci, and 0.99 with ten diagnostic loci. The proportion of rainbow trout admixture in each population was calculated as the number of rainbow trout alleles divided by total number of alleles genotyped.

Linear models and Akaike’s Information Criterion (AIC) were used to test for relationships between variables hypothesized to influence introgression and to select the best-supported models. Specifically, we tested whether biotic and climatic variation were related to the amount of rainbow trout hybridization across sites (N = 50) and over time on the basis of repeat samples from the late 1970s and early 1980s and early 2000s (N = 20 sites; the source population was not included in the model). Before spatial model-selection analyses, proportions of rainbow trout introgression (p) were adjusted by ((p(N − 1) + 1/2)/N) to avoid zeros, then logit transformed to linearize relationships. The response variable for temporal change in hybridization was the observed proportion of introgression in recent samples minus the proportion in historical samples. To determine if climatic effects were still evident after removing the effects of distance to source, we tested for relationships between temperature and/or precipitation and residuals from the relationship between distance to source and proportion of introgression.

Predictor variables for each population included flow-connected stream distances to the source of rainbow trout5, 20, 30, spring precipitation (April, May, June, and three month average), summer stream temperature, and the presence or absence of recent wildfire (since 1984) within each stream drainage. We tested for additive and interactive effects between covariates. Covariates were obtained for each location using ArcGIS version 10.1 (Environmental Systems Research Institute, Redlands). Average summer stream temperature conditions were predicted using a spatially explicit stream temperature model for the Flathead River basin26 driven by high-resolution air temperature surfaces (800 m). Average precipitation conditions were calculated from daily precipitation surfaces (1 km) processed from National Aeronautics and Space Administration (NASA) Daymet data. Because genetic samples were collected from multiple age classes (age−1–age−3), climatic covariates were averaged over the three years prior to genetic sampling. Average precipitation for the month of May was found to have the strongest statistical relationship with introgression of all spring months and average conditions tested. Therefore, the temporal analyses used the change in average stream temperature and May precipitation from 1980–1984 to 2000–2005 as predictor variables.

All model results (Supplementary Tables 2 and 4), parameter estimates from best-supported models (Supplementary Table 3), and correlations between predictor variables (Supplementary Table 5) can be found in the Supplementary Methods.

  1. Parmesan, C. Ecological and evolutionary responses to recent climate change. Ann. Rev. Ecol. Evol. Syst. 37, 637669 (2006).
  2. Hoffmann, A. A. & Sgro, C. M. Climate change and evolutionary adaptation. Nature 470, 479485 (2011).
  3. Moritz, C. & Agudo, R. The future of species under climate change: Resilience or decline? Science 341, 504508 (2013). URL:
http://www.nature.com/nclimate/journal/v4/n7/full/nclimate2252.html
Citation statistics:
资源类型: 期刊论文
标识符: http://119.78.100.158/handle/2HF3EXSE/5127
Appears in Collections:气候变化事实与影响
科学计划与规划
气候变化与战略

Files in This Item: Download All
File Name/ File Size Content Type Version Access License
nclimate2252.pdf(1253KB)期刊论文作者接受稿开放获取View Download

Recommended Citation:
Clint C. Muhlfeld. Invasive hybridization in a threatened species is accelerated by climate change[J]. Nature Climate Change,2014-05-25,Volume:4:Pages:620;624 (2014).
Service
Recommend this item
Sava as my favorate item
Show this item's statistics
Export Endnote File
Google Scholar
Similar articles in Google Scholar
[Clint C. Muhlfeld]'s Articles
百度学术
Similar articles in Baidu Scholar
[Clint C. Muhlfeld]'s Articles
CSDL cross search
Similar articles in CSDL Cross Search
[Clint C. Muhlfeld]‘s Articles
Related Copyright Policies
Null
收藏/分享
文件名: nclimate2252.pdf
格式: Adobe PDF
此文件暂不支持浏览
所有评论 (0)
暂无评论
 

Items in IR are protected by copyright, with all rights reserved, unless otherwise indicated.