Determining puma habitat suitability in the Eastern USA

The largemouth bass Micropterus salmoides is arguably the most sought fish species by inland anglers in North America. The suitability of inland habitats for supporting largemouth bass populations can differ remarkably but is difficult to quantify. I developed and evaluated the utility of a habitat suitability index (HSI) for largemouth bass in tidal freshwater portions of selected drainages in the Chesapeake Bay. The HSI comprised suitability scores ranging from 0 (not suitable) to 1 (perfectly suitable) and was calculated from seven separate suitability models in which suitability scores were functions of summer water temperature (°C), summer dissolved oxygen (mg/L), summer pH, maximum monthly salinity (‰), submerged aquatic vegetation coverage (%), monthly average water clarity, and stream discharge (m3/s) during the spawning season. Some of the suitability models that contributed to the HSI were modified to reflect species–habitat relationships observed for largemouth bass in drainages of the Chesapeake Bay from 2003 to 2008. Based on the HSI, drainages ranged from moderately suitable (HSI = 0.57) to highly suitable (HSI = 0.84). Spatial and monthly differences in HSI were positively correlated with relative abundance of largemouth bass; similar habitat conditions among years led to low interannual variation in HSI, which was not correlated with annual differences in relative abundance. Although the HSI may be useful in summarizing habitat suitability for a species, it may obscure important physical and biotic interactions that could affect the species’ distribution.

IntroductionElucidating the relationship between habitat characteristics and population parameters is critical for effective conservation. Habitat suitability index (HSI) models are often used in wildlife management and conservation practice assuming that they predict species occurrence, abundance and demography. However, the relationship between vital rates such as survival and reproduction and habitat suitability has rarely been evaluated. In this study, we used pond occupancy and mark-recapture data to test whether HSI predicts occupancy, reproduction and survival probabilities. Our model species is the great crested newt (Triturus cristatus), a pond-breeding amphibian protected under the European Habitats Directive.ResultsOur results show a positive relationship between the HSI and reproduction probability, whereas pond occupancy and survival probabilities were not related to HSI. Mortality was found to be higher during breeding seasons when newts are in ponds than during terrestrial phases of adult newts.ConclusionHabitat suitability models are increasingly applied to wildlife management and conservation practice. We found that the HSI model predicted reproduction probability, rather than occurrence or survival. If HSI models indicate breeding populations rather than mere species occurrences, they may be used to identify habitats of higher priority for conservation. Future HSI models might be improved through modelling breeding populations vs. non-breeding populations rather than presence/absence data. However, according to our results the most suitable habitat is not necessarily the habitat where demographic performance is best. We recommend that conservation practitioners should use HSI models cautiously because there may be no direct link between habitat suitability, demography and consequently, population viability.Electronic supplementary materialThe online version of this article (doi:10.1186/s12983-015-0103-3) contains supplementary material, which is available to authorized users.

The empirical habitat suitability index (HSI) has been widely used to examine the habitat characteristics of terrestrial animals, though rarely used in highly migratory fish such as tuna. This study used the geographic information system technique to establish empirical models of HSI for yellowfin tuna (YFT) in the Western and Central Pacific Ocean (WCPO). Daily catch data from the Taiwanese purse seine fishery during 2003–2007 were aggregated monthly into sequential degrees before match processing the conducted data to obtain monthly remote-sensing data for multi-environmental factors, including sea surface temperature (SST), chlorophyll-a (chl-a), sea surface height (SSH) and sea surface salinity (SSS). According to the frequency distribution of each factor on which YFT were caught, this study transformed the values of the four factors into a suitability index (SI) ranging from low to high (0–1). These SI values were consequently combined into different empirical HSI models, and the optimum models were selected using the general linear model. The optimum empirical HSI for YFT in the study area was converted for SI (SST, SSH, chl-a and SSS) using the arithmetic mean model, of which the correct prediction rate was 71.9%. An agreement was present between the average HSI and total YFT catch. Furthermore, the high HSI area corresponds with the displacement of catch per unit effort (CPUE).

Historical and continuing US Department of Energy (USDOE) operations have not only contaminated the environment but have altered the landscape as well as habitat quality and quantity. The changes in the landscape that have occurred within USDOE facilities may be as important as assessing the effects resulting from the release of chemical contaminants into the environment. We examined several methods from habitat analysis and landscape ecology for their possible application to ecological risk assessment at selected USDOE facilities. In an earlier manuscript, we presented landscape ecological methods that could be used in ecological risk assessment. In this manuscript, we present results of adapting Habitat Suitability Index (HSI) models, developed for habitat suitability analysis, for use in ecological risk assessment at USDOE facilities. The Oak Ridge Reservation (ORR) in eastern Tennessee and the Savannah River Site (SRS) in western South Carolina were selected as representative USDOE facilities. HSI models were developed in a probabilistic format for species common to both facilities [white-tailed deer (Odocoileus virginianus) and pileated woodpecker (Dryocopus pileatus)] and species exclusive to each facility [red-spotted newt (Notophthalmus viridescens viridescens) for the ORR and American alligator (Alligator mississippiensis) for the SRS]. Results indicate that HSI models could effectively be used to assess ecological risks. Spatially-explicit HSI models and landscape-scale ecological models that provide the necessary spatial component for more realistic risk assessments that could be used in ecological risk assessments at a variety of federal facilities, such as USDOE facilities, are presented as a result of a literature review and an internet search. The combination and integration of methods from habitat suitability analysis and ecological models presented in this manuscript and methods from landscape ecology presented in an earlier manuscript represent an effective set of tools that could be used to assess ecological risks at USDOE facilities; these tools could also be used at other locations.

Sea level rise, amplified by anthropogenic climate change, causes visible impacts on wildlife habitats in low-lying coastal areas. However, the potential consequences of sea level height (SLH) changes for pelagic fish habitats in the open oceans are poorly understood, especially for climate-sensitive Ommastrephid squid. In this study, the impacts of SLH changes were estimated under three different SLH change scenarios (5 cm, 20 cm, and 35 cm) for the western stock of winter-spring neon flying squid Ommastrephes bartramii, an ecologically- and commercially-important species in the northwest Pacific Ocean (NWPO). A habitat suitability index (HSI) modeling approach was applied to predict the habitat suitability of O. bartramii, using data from 2006 to 2014 for model construction and data from 2015 for model validation. Results showed that the decreasing catch-per-unit-effort (CPUE) of O. bartramii from 2006 to 2015 was highly correlated with the increases in sea level height (SLH). Significant positive relationships were found between the HSI values for latitudinal centers of gravity (LATGHSI) and SLH. The monthly-averaged HSI of O. bartramii was negatively correlated with the SLH, except in July and November. SLH scenarios revealed that the predicted suitable habitat (areas with HSI≥0.6) exhibited a decreasing trend with increasing SLH from August to October. In July and November, the suitable habitat initially increased in the +5 cm and +20 cm scenarios, but largely declined under the +35 cm scenario. The poor habitat (areas with HSI≤0.2) significantly increased under all SLH scenarios. A poleward shift of LATGHSI was observed under all scenarios. These findings suggest that the O. bartramii habitats in the NWPO are at risk to SLH changes and have important implications for better conservation and fishery management regarding pelagic short-lived Ommastrephid squid in global oceans.

A habitat suitability index (HSI) model for rock bass Ambloplites rupestris was developed from available suitability curves and field tested by comparing HSI and habitat units with seasonal standing stock, biomass, and production of rock bass in two fourth-order streams in Virginia. Consistent ranking between HSI and mean standing stocks for the study year was observed only for 250-m stream sections, the largest spatial unit of comparison. In contrast, changes in HSI explained no more than 22% of the variability in mean standing stocks in 50-m subsections of Back Creek, and were not significantly correlated with standing stocks in Little Walker Creek. Factors that apparently contributed to low correlations and ranking success included geographically homogeneous habitat and seasonal fish movements, as well as potentially important variables such as depth and cover that are not features of the tested HSI models. Annual discharge and fish distribution data further suggested that physical habitat was more limiting to populations of rock bass in the fall, not during the summer low-flow period as previously assumed. Unmeasured temporal variability in the environment may have been equally important in limiting presumed carrying capacities in study streams. Annual production for rock bass was nearly identical in both streams and averaged 1.04 g˙m−2˙year−1. This similarity in annual production was consistent with the habitat homogeneity implied by the narrow range of observed HSIs. Validation of our HSI model and similar short-term assessments of habitat quality would be more likely if study sections included the home range or annual habitat requirements of all life stages of the target species. Additional quantitative data on species-habitat relationships and systematic evaluation of individual suitability indices should also improve testing of aggregate HSI models.

Sustainable use of fishery resources requires the effective monitoring and managing of fish stocks and fish habitats. Chub mackerel (Scomber japonicus), distributed in the East China Sea and Yellow Sea, are mainly caught by purse seine fishing fleets from China, Japan, and South Korea. This study used fishery data from Chinese large lighting–purse seine fleets and environmental data including sea surface height (SSH), sea surface temperature (SST) from remote sensing, and temperature gradient derived from SST (GSST) during 1998–2010 to develop habitat suitability index (HSI) models. The HSI models were then used to identify hotspots for chub mackerel for each month. HSI models were developed separately for each of the three distribution areas defined for chub mackerel. According to the frequency distribution of the fishing effort with respect to three environmental variables, suitability index (SI) values were calculated and SI models for each environmental variable were established. The three SI models were combined into two different empirical HSI models: the arithmetic mean model (AMM) and the geometric mean model (GMM). The results showed that the AMM was more suitable than the GMM to estimate the HSI for chub mackerel. The monthly latitudinal variation trend of hotspots was consistent with that of the gravity centres of fishing effort in almost all months. Hotspot maps based on the predicted HSI values were validated by fishery data in 2011. This result indicates that the HSI model based on the AMM can reliably predict hotspots for chub mackerel in the coastal waters of China.

A widely used system for assessing habitat for the great crested newt uses five categories ranging from ‘poor’ to ‘excellent’ based on thresholds for the Habitat Suitability Index (HSI). However, how these categories relate to pond occupancy, at an England-wide scale, is unknown. Equally, the Habitat Suitability Index system has so far only been validated using traditional direct observation methods rather than environmental DNA protocols that are becoming commonplace. Without further validation on a national scale, misleading decisions may be made concerning the likely presence or likely absence of great crested newts. Using environmental DNA data collected from over 5300 ponds distributed across much of England, we show that the existing scoring system underestimates pond occupancy in the lower categories and overestimates pond occupancy in the higher categories, while the median habitat suitability index value was found just within the 'good' category. We found that the median habitat suitability index for occupied ponds was 0.7, confirming this value as a target to aim for when creating or restoring ponds for great crested newts. We suggest a revised system based on the median occupied pond HSI score, whereby the two extreme 'poor' and 'excellent' categories each contain just 10 % of occupied ponds; the 'below average' and 'good' categories each contain 20 % of all occupied ponds, and the 'average' category contains the central 40 % of occupied ponds. Although regional variation in estimated pond occupancy rates using this system may need to be accounted for when interpreting HSI scores, the revised scoring system is generally robust across England. Both the existing and revised HSI scoring systems are no substitute for surveys, and caution is needed when interpreting absence of newts based on habitat suitability data only. Keywords: Habitat Suitability Index, Triturus cristatus, HSI, UK regions, environmental DNA

The Florida sandhill crane Antigone canadensis pratensis is designated as threatened by the state of Florida, where there is an urgent need to map and quantify available habitat. First, we used habitat suitability index (HSI) modelling to map and assess potential nesting habitat for sandhill cranes in Florida. Second, we used spatial optimization approaches to calculate the maximum number of breeding pairs that can simultaneously occupy potential nests given that they both must be of some minimum quality and must be spaced some minimal distance apart. Mapping results reveal that nesting habitat is concentrated in the central portion of the state, with adequate brooding habitat appearing to be the most limiting factor affecting habitat suitability. Assuming nesting only occurs in habitat rated as high quality (HSI ≥ 0.7) and spacing between adjacent nests is ≥1,000 m, we conservatively estimate that 5,540 nesting pairs of Florida sandhill cranes can potentially be supported. Additional nesting pairs may be supported in habitats of marginal (HSI ≥ 0.3; 14,530) to moderate (HSI ≥ 0.5; 8,723) quality. The suitability maps and breeding pair estimates can be used to identify important habitat areas to focus crane conservation efforts, determine potentially limiting habitat features across the landscape, and potentially guide future population monitoring efforts. For example, grassland or prairie restoration could be used to potentially increase nesting pairs in the southern portion of the state where emergent wetlands are abundant but brooding habitat is lacking.

A Habitat Suitability Index (HSI) model was developed using four environmental variables (bottom temperature, bottom salinity, depth, and bottom substrate type) and 29 years of spring and fall lobster surveys for evaluating the spatiotemporal variability of suitable lobster habitat in Long Island Sound (LIS). The suitability indices calculated for the four environmental variables were combined to form a composite HSI using an arithmetic mean model and geometric mean model. A cross-validation study was conducted to evaluate the predictive performance of the HSI models. Annual geographic information system maps of estimated HSI values were produced using Kriging interpolation for adult and juveniles in spring and fall from 1978 to 2012. The overall spatial distribution of suitable habitat for lobster was mainly concentrated in the western-central part of LIS during spring (April–June), but showed clustering patterns throughout LIS during fall (September–October). An examination of the temporal change in annual median HSI values identified possible time blocks when habitat conditions were extremely poor and revealed a statistically significant decreasing trend in availability of suitable habitat for juveniles during spring from 1978 to 2012. Spatiotemporal variability in availability of suitable habitat may imply changes in carrying capacity of LIS for the American lobster.

ABSTRACTIn Mediterranean farming landscapes, natural waters are declining, leading to serious consequences for amphibians. In this context, artificial water bodies, used for livestock supply, may represent an ecological surrogate for natural ones. Here, we investigated the habitat suitability of natural and artificial aquatic sites for Triturus carnifex within a farming area where traditional practices are declining. The main goal was to clarify if artificial sites represent keystone habitats contributing to amphibian persistence. Site suitability was assessed by a Habitat Suitability Index (HSI), a body condition index (BCI) and evaluating newt trophic strategy in terms of diet composition and individual specialization (V). HSI was similar in artificial and natural waters, with most study sites falling into the “Good”‐“Average” categories. Drinking troughs seem to fulfill the ecological requirements of the species with a habitat suitability that can be improved by a proper management. BCI did not vary between artificial and natural sites, suggesting that newts were well locally adapted, exploited resources with similar energetic cost‐efficiency, and experienced similar predation and competition. Diet composition varied between artificial and natural sites because of the opportunistic feeding behavior of T. carnifex, the specialization towards different prey, and site‐specific environmental conditions. Individual specialization was significant in artificial and natural sites, confirming the availability of resources in drinking troughs. Indeed, V is more frequent in environments where prey is highly available and profitable to hunt. Our findings suggest that artificial sites with specific habitat suitability sustain specialized individuals with good body conditions. If properly managed, they may support newt vital activities, especially feeding, representing an ecological surrogate for natural sites. In the face of freshwater loss, they represent an opportunity to preserve amphibians in farming landscapes. The recognition of their conservation value is timely, given the spread of intensive farming and the threatened status of many amphibians.

Ocean warming-induced range-shifting of potential habitat for jumbo flying squid Dosidicus gigas in the Southeast Pacific Ocean off Peru

Chang, Y.-J., Sun, C.-L., Chen, Y., Yeh, S.-Z., DiNardo, G., and Su, N.-J. 2013. Modelling the impacts of environmental variation on the habitat suitability of swordfish, Xiphias gladius, in the equatorial Atlantic Ocean. – ICES Journal of Marine Science, 70: 1000–1012. We developed a habitat suitability index (HSI) model to identify the optimal habitats of swordfish (Xiphias gladius) in the equatorial Atlantic Ocean. Environmental variables, including sea surface temperature (SST), mixed layer depth (MLD), chlorophyll-a concentrations, and sea surface height anomaly, as well as catch and effort data from Taiwanese longline fisheries, were used. The geometric mean model including all the above environmental variables was identified as the most parsimonious model for yielding HSI predictions coinciding with productive fishing grounds with high fishing effort. Swordfish mainly aggregated in the northwest region during March–May and spread southeast thereafter in response to seasonal shifts in oceanographic conditions. There was annual variation in the distribution of habitat patches, and the habitat quality was reduced in the northwest region of the equatorial Atlantic Ocean during 2005. The apparent spatial shifts in optimal habitats might be linked to reduced MLD and elevation in sea surface height, which might be related to climate variability (e.g. Niño-Southern Oscillation and/or Northern Atlantic Oscillation). Because environmental data regarding climate change scenarios are becoming readily available, we can utilize the proposed HSI models to evaluate possible changes in habitat suitability resulting from climate change and provide scientific advice for the development of management regulations.

Species-specific habitat suitability models have potential for use in restoration efforts, but their efficiency still remains in question. As eastern brook trout, Salvelinus fontinalis, populations in New England have declined over the last few decades as a result of anthropogenic and natural impacts on critical habitat, habitat suitability models have become a common tool for conservation efforts. These models, however, have inherent flaws that prevent widespread and uniform use. To better adapt these models, the flaws must be properly addressed. Using a modified habitat suitability index (HSI) model developed for eastern brook trout, we explained the correlation between catch per unit effort (CPUE) of brook trout with designated HSI variables such as temperature, dominant substrate type, and percent riffle fines for nineteen reaches within the Westfield River watershed. CPUE was not significantly correlated with HSI outputs. A principal components analysis (PCA) was employed and revealed driving factors within the system. Four variables were shown to yield the highest explained variance over the first two axes: velocity, instream cover, percent pools, and thalweg depth. Evidence suggests that habitat assessment based around these core variables may lead to a more efficient and accurate assessment. Recommendations for improved methodologies include revised tolerance curves, a reworked index rating system, and revised model variables based on current field research. Alterations to existing models provide hope for more accurate assessment, and increased efficiency in conservation efforts.

Understanding relationships between habitat associations for individuals and habitat factors that limit populations is a primary challenge for managers of stream fishes. Although habitat use by individuals can provide insight into the adaptive significance of selected microhabitats, not all habitat parameters will be significant at the population level, particularly when distributional patterns partially result from habitat degradation. We used underwater observation to quantify microhabitat selection by an imperiled stream fish, the Candy Darter Etheostoma osburni, in two streams with robust populations. We developed multiple‐variable and multiple‐life‐stage habitat suitability indices (HSIs) from microhabitat selection patterns and used them to assess the suitability of available habitat in streams where Candy Darter populations were extirpated, localized, or robust. Next, we used a comparative framework to examine relationships among (1) habitat availability across streams, (2) projected habitat suitability of each stream, and (3) a rank for the likely long‐term viability (robustness) of the population inhabiting each stream. Habitat selection was characterized by ontogenetic shifts from the low‐velocity, slightly embedded areas used by age‐0 Candy Darters to the swift, shallow areas with little fine sediment and complex substrate, which were used by adults. Overall, HSIs were strongly correlated with population rank. However, we observed weak or inverse relationships between predicted individual habitat suitability and population robustness for multiple life stages and variables. The results demonstrated that microhabitat selection by individuals does not always reflect population robustness, particularly when based on a single life stage or season, which highlights the risk of generalizing habitat selection that is observed during nonstressful periods or for noncritical resources. These findings suggest that stream fish managers may need to be cautious when implementing conservation measures based solely on observations of habitat selection by individuals and that detailed study at the individual and population levels may be necessary to identify habitat that limits populations.Received March 30, 2016; accepted July 19, 2016Published online October 7, 2016