Movement distances enhance validity of predictive models

Including the distance species are able to move in predictive models improves conservation practice. Bird inventory projects carried out from 1993 to 2004 in Taiwan provide an opportunity to investigate the relationships among species distribution, movement distance, and the environment. We compared projected distributions of 17 Taiwanese endemic bird species using what we called the Standard Method (i.e. movement distance is zero) and what we called the Buffer Method (i.e. movement distance is longer than zero) in three presence-only models (GARP, MAXENT and LIVES). The Standard Method used species original occurrence records directly while the Buffer Method expanded the occurrence of species to areas 1 km² around each recorded location. We first tested the efficacy of the Buffer Method using ten common species of the 17, and then applied the method to two rare species of the 17. For both the common and rare species, the distributions predicted by the two methods showed slight but important differences. The Buffer Method for all species had a higher average predictive probability, while the Standard Method had a higher maximum predictive probability. Most of the values for the area under the curve (AUC) were over 0.8 with the exceptions of Taiwan Barbet (Megalaima nuchalis) and Taiwan Hwamei (Garrulax taewanus), which have recently separated from Indochinese Barbet (Megalaima annamensis) and Chinese Hwamei (Garrulax canorus), and since 2008 and 2006 have been regarded as species endemic to the study area. Kappa values showed good performance for all species using both methods. The Buffer Method, however, resulted in significantly higher sensitivity and accuracy values for all models of species (p < 0.05). We conclude that when modeling species distribution including the area where the species was censused along with areas within the minimum movement areas better defines the surrounding areas that might supplement core habitat requirements. Therefore, using the Buffer Method, species surrounding distribution can be obtained which provides a better understanding of the species distributions. Given that distribution size is a key to the conservation of species, we suggest the Buffer Method can be used in conservation planning.     

Influence of grain size on species-habitat models

High resolution remote sensing data facilitate the use of small-scale habitat features such as trees or hedges in the analysis of species-habitat relationships. Such data potentially enable more accurate species-habitat mapping than lower resolution data. Here, for the first time, we systematically investigated this hypothesis by altering the spatial resolution from 1 m up to 1000 m grain size in species-habitat models of 13 bird species. The study area covered the Nidda river catchment in central Germany, a large heterogeneous landscape of 1620 km². A high resolution habitat map of the area was converted to coarser spatial and thematic resolutions in seven steps. We investigated how model performance responded to grain size, and we compared the differential effects of spatial resolution and thematic resolution on model performance. Explained deviance (D²) of the bird models generally decreased with coarser spatial resolution of the data, although it did not decrease monotonically in all species. On average across all species, model D² decreased from 41.5 at 1 m grain size to 15.9 at 1000 m grain size. Ten species were best modelled at 1 m, two species at 3 m and one species at 32 m grain size. Model performance degraded continuously with increasing grain size, both in habitat generalist and habitat specialist bird species, and was systematically lower in habitat generalists. The higher model performance observed at finer grain sizes was most likely caused by the combination of three factors: (1) high spatial accuracy of bird records and (2) a more precise location and delineation of habitat features and, (3) to a lesser degree, by more habitat types differentiated in maps of finer resolution. We conclude that higher spatial and thematic resolution data can be essential for deriving accurate predictions on bird distribution patterns from species-habitat models. Especially for bird species that are sensitive to specific land-use types or to small-scaled habitat features, a grain size of 1-3 m seems most promising.     

Incorporating low-resolution historic species location data decreases performance of distribution models

Developing robust species distribution models is important as model outputs are increasingly being incorporated into conservation policy and management decisions. A largely overlooked component of model assessment and refinement is whether to include historic species occurrence data in distribution models to increase the data sample size. Data of different temporal provenance often differ in spatial accuracy and precision. We test the effect of inclusion of historic coarse-resolution occurrence data on distribution model outputs for 187 species of birds in Australian tropical savannas. Models using only recent (after 1990), fine-resolution data had significantly higher model performance scores measured with area under the receiver operating characteristic curve (AUC) than models incorporating both fine- and coarse-resolution data. The drop in AUC score is positively correlated with the total area predicted to be suitable for the species (R² = 0.163-0.187, depending on the environmental predictors in the model), as coarser data generally leads to greater predicted areas. The remaining unexplained variation is likely to be due to the covariate errors resulting from resolution mismatch between species records and environmental predictors. We conclude that decisions regarding data use in species distribution models must be conscious of the variation in predictions that mixed-scale datasets might cause.     

Maximum entropy modeling of species geographic distributions

The availability of detailed environmental data, together with inexpensive and powerful computers, has fueled a rapid increase in predictive modeling of species environmental requirements and geographic distributions. For some species, detailed presence/absence occurrence data are available, allowing the use of a variety of standard statistical techniques. However, absence data are not available for most species. In this paper, we introduce the use of the maximum entropy method (Maxent) for modeling species geographic distributions with presence-only data. Maxent is a general-purpose machine learning method with a simple and precise mathematical formulation, and it has a number of aspects that make it well-suited for species distribution modeling. In order to investigate the efficacy of the method, here we perform a continental-scale case study using two Neotropical mammals: a lowland species of sloth, Bradypus variegatus, and a small montane murid rodent, Microryzomys minutus. We compared Maxent predictions with those of a commonly used presence-only modeling method, the Genetic Algorithm for Rule-Set Prediction (GARP). We made predictions on 10 random subsets of the occurrence records for both species, and then used the remaining localities for testing. Both algorithms provided reasonable estimates of the species’ range, far superior to the shaded outline maps available in field guides. All models were significantly better than random in both binomial tests of omission and receiver operating characteristic (ROC) analyses. The area under the ROC curve (AUC) was almost always higher for Maxent, indicating better discrimination of suitable versus unsuitable areas for the species. The Maxent modeling approach can be used in its present form for many applications with presence-only datasets, and merits further research and development.     

Modeling a spatially restricted distribution in the Neotropics: How the size of calibration area affects the performance of five presence-only methods

We here examine species distribution models for a Neotropical anuran restricted to ombrophilous areas in the Brazilian Atlantic Forest hotspot. We extend the known occurrence for the treefrog Hypsiboas bischoffi (Anura: Hylidae) through GPS field surveys and use five modeling methods (BIOCLIM, DOMAIN, OM-GARP, SVM, and MAXENT) and selected bioclimatic and topographic variables to model the species distribution. Models were first trained using two calibration areas: the Brazilian Atlantic Forest (BAF) and the whole of South America (SA). All modeling methods showed good levels of predictive power and accuracy with mean AUC ranging from 0.77 (BIOCLIM/BAF) to 0.99 (MAXENT/SA). MAXENT and SVM were the most accurate presence-only methods among those tested here. All but the SVM models calibrated with SA predicted larger distribution areas when compared to models calibrated in BAF. OM-GARP dramatically overpredicted the species distribution for the model calibrated in SA, with a predicted area around 10⁶ km² larger than predicted by other SDMs. With increased calibration area (and environmental space), OM-GARP predictions followed changes in the environmental space associated with the increased calibration area, while MAXENT models were more consistent across calibration areas. MAXENT was the only method that retrieved consistent predictions across calibration areas, while allowing for some overprediction, a result that may be relevant for modeling the distribution of other spatially restricted organisms.     

The derivation of species response curves with Gaussian logistic regression is sensitive to sampling intensity and curve characteristics

We investigated quantitatively the sensitivity of plant species response curves to sampling characteristics (number of plots, occurrence and frequency of species), along a simulated pH gradient. We defined 54 theoretical unimodal response curves, issued from combinations of six values for optimum (opt = 3, 4, …, 8), three values for tolerance (tol = 0.5, 1.0, and 1.5, sensu ter Braak and Looman [ter Braak, C.J.F., Looman, C.W.N., 1986. Weighted averaging, logistic regression and the Gaussian response model. Vegetatio 65, 3–11]), and three values for maximum probability of presence (p_max = 0.05, 0.20, and 0.50). For each of these 54 theoretical response curves, we built artificial binary data sets (presence/absence) to test the influence of species occurrence, frequency, or number of available plots. With real data extracted from EcoPlant, a phytoecological database for French forests [Gégout, J.-C., Coudun, Ch., Bailly, G., Jabiol, B., 2005. EcoPlant: a forest sites database linking floristic data with soil characteristics and climatic conditions. J. Veg. Sci. 16, 257–260], we compared the ecological response of 50 plant species to soil pH, based first on a small data set (100 randomly sampled plots), and then based on the whole data set available (3810 plots).     

Obtaining Environmental Favourability Functions from Logistic Regression

Logistic regression is a statistical tool widely used for predicting species’ potential distributions starting from presence/absence data and a set of independent variables. However, logistic regression equations compute probability values based not only on the values of the predictor variables but also on the relative proportion of presences and absences in the dataset, which does not adequately describe the environmental favourability for or against species presence. A few strategies have been used to circumvent this, but they usually imply an alteration of the original data or the discarding of potentially valuable information. We propose a way to obtain from logistic regression an environmental favourability function whose results are not affected by an uneven proportion of presences and absences. We tested the method on the distribution of virtual species in an imaginary territory. The favourability models yielded similar values regardless of the variation in the presence/absence ratio. We also illustrate with the example of the Pyrenean desman’s (Galemys pyrenaicus) distribution in Spain. The favourability model yielded more realistic potential distribution maps than the logistic regression model. Favourability values can be regarded as the degree of membership of the fuzzy set of sites whose environmental conditions are favourable to the species, which enables applying the rules of fuzzy logic to distribution modelling. They also allow for direct comparisons between models for species with different presence/absence ratios in the study area. This makes them more useful to estimate the conservation value of areas, to design ecological corridors, or to select appropriate areas for species reintroductions. Received: June 2005 / Revised: July 2005     

Predicting tree species presence and basal area in Utah: A comparison of stochastic gradient boosting,generalized additive models,and tree-based methods

Many efforts are underway to produce broad-scale forest attribute maps by modelling forest class and structure variables collected in forest inventories as functions of satellite-based and biophysical information. Typically, variants of classification and regression trees implemented in Rulequest's^© See5 and Cubist (for binary and continuous responses, respectively) are the tools of choice in many of these applications. These tools are widely used in large remote sensing applications, but are not easily interpretable, do not have ties with survey estimation methods, and use proprietary unpublished algorithms. Consequently, three alternative modelling techniques were compared for mapping presence and basal area of 13 species located in the mountain ranges of Utah, USA. The modelling techniques compared included the widely used See5/Cubist, generalized additive models (GAMs), and stochastic gradient boosting (SGB). Model performance was evaluated using independent test data sets. Evaluation criteria for mapping species presence included specificity, sensitivity, Kappa, and area under the curve (AUC). Evaluation criteria for the continuous basal area variables included correlation and relative mean squared error. For predicting species presence (setting thresholds to maximize Kappa), SGB had higher values for the majority of the species for specificity and Kappa, while GAMs had higher values for the majority of the species for sensitivity. In evaluating resultant AUC values, GAM and/or SGB models had significantly better results than the See5 models where significant differences could be detected between models. For nine out of 13 species, basal area prediction results for all modelling techniques were poor (correlations less than 0.5 and relative mean squared errors greater than 0.8), but SGB provided the most stable predictions in these instances. SGB and Cubist performed equally well for modelling basal area for three species with moderate prediction success, while all three modelling tools produced comparably good predictions (correlation of 0.68 and relative mean squared error of 0.56) for one species.     

Mapping eucalypt forest susceptible to dieback associated with bell miners (Manorina melanophys) using laser scanning, SPOT 5 and ancillary topographical data

Mapping the location and extent of forest at risk from damaging agents or processes assists forest managers in prioritizing their planning and operational mitigation activities. In Australia, Bell Miner Associated Dieback (BMAD) refers to a form of canopy decline observed in eucalypt crowns occupied by colonies of bell miners (Manorina melanophrys). High densities of bell miners are associated with decreased avian abundance and diversity and an increase in psyllid abundance on crown foliage. BMAD has recently been nominated as a key threatening process in New South Wales (NSW). Consequently, a modelling system for predicting bell miner distribution in coastal eucalypt forests of NSW has been developed. The presence or absence of bell miners was recorded in 130 plots located within a 12,800 ha catchment study area containing a range of eucalypt forest types. The modelling system was produced by integrating a machine learning software suite (WEKA), and the statistical software R within the geographic resources analysis support system (GRASS) geographical information system (GIS). The variable modelled was the binary variable: presence or absence of bell minors. Six modelling techniques (Logistic regression; generalised additive models; two tree-based ensemble classification algorithms, random forest and Adaboost and Neural Networks) were integrated with airborne laser scanning; SPOT 5 and topographic derived variables. Model evaluation and parameter selection were measured by three threshold dependent measures (sensitivity, specificity and kappa) and the threshold independent Receiver Operator Curve (ROC) analysis. The final presence and absence maps were obtained through maximisation of the kappa statistic and applied at a resolution of 10 m across the entire catchment study area. For this data set, the most accurate algorithm for predicting the distribution of bell miner colonies was random forest (kappa = 0.84; ROC area under curve = 0.97). Variables most commonly selected in the six models were the laser scanning metrics; coefficient of variation, skewness, and the 10th and 90th percentiles derived from the shape of the height frequency distribution which, in turn, is directly influenced by vertical structure of the forest. An image textural statistic based on the shortwave infrared (SWIR) band of SPOT 5 was also commonly selected by the models. The SWIR band is sensitive to vegetation and soil moisture content. These models predicted that forest stands with a sparse eucalypt canopy over a moist, dense understorey were susceptible to being colonised by bell miners and hence BMAD.     

Temporal limits to simulating the future spread pattern of invasive species: Buddleja davidii in Europe and New Zealand

The spread of invasive species is a major ecological and economic problem. Dynamic spread modelling is a potentially valuable tool to assist regional and central government authorities to monitor and control invasive species. To date a lack of suitable data has meant that most broad scale dispersal models have not been validated with independent datasets, and so their predictive ability and reliability has remained unscrutinised. A dynamic, stochastic dispersal model of the widely invasive plant Buddleja davidii was calibrated on European spread data and then used to project the temporal progression of B. davidii's distribution in New Zealand, starting from several different historical distributions. To assess the model's performance, we constructed an occupancy map based on the average number of simulation realisations that have a population present. The application of Receiver Operating Characteristic (ROC) curves to occupancy maps is introduced, but with specificity substituted by the proportion of available area used in a realisation. A derivative measure, the partial area under these curves when assessed through time (pAUC), is introduced and used to assess overall performance of the spread model. The model was able to attain a high level of model sensitivity, encompassing all of the known locations within the occupancy envelope. However, attempting to simulate the spread of this invasive species beyond a decade had very low model specificity. This is due to several factors, including the exponential process of spread (the further a population spreads the more sites exist from which it can spread stochastically), and the Markovian chain property of the stochastic system whereby differences between realisations compound through time. These features are seen in many reports of spread models, without being explicitly acknowledged. Our measure of pAUC through time allows a model's temporal performance and its specificity to be simultaneously assessed. While the rapid deterioration in model performance limits the utility of this type of modelling for forecasting long-term broad-scale strategic management of biological invasions, it does not necessarily limit its attractiveness for informing smaller scale and shorter term invasion management activities such as surveillance, containment and local eradication.     

Integrating habitat-masked range maps with quantifications of prevalence to estimate area of occupancy in IUCN assessments

Estimates of species geographic ranges constitute critical input for biodiversity assessments, including those for the International Union for the Conservation of Nature (IUCN) Red List of Threatened Species. Area of occupancy (AOO) is one metric that IUCN uses to quantify a species’ range, but data limitations typically lead to either under- or overestimates (and unnecessarily wide bounds of uncertainty). Fortunately, existing methods in which range maps and land-cover data are used to estimate the area currently holding habitat for a species can be extended to yield an unbiased range of plausible estimates for AOO. Doing so requires estimating the proportion of sites (currently containing habitat) that a species occupies within its range (i.e., prevalence). Multiplying a quantification of habitat area by prevalence yields an estimate of what the species inhabits (i.e., AOO). For species with intense sampling at many sites, presence–absence data sets or occupancy modeling allow calculation of prevalence. For other species, primary biodiversity data (records of a species’ presence at a point in space and time) from citizen-science initiatives and research collections of natural history museums and herbaria could be used. In such cases, estimates of sample prevalence should be corrected by dividing by the species’ detectability. To estimate detectability from these data sources, extensions of inventory-completeness analyses merit development. With investments to increase the quality and availability of online biodiversity data, consideration of prevalence should lead to tighter and more realistic bounds of AOO for many taxonomic groups and geographic regions. By leading to more realistic and representative characterizations of biodiversity, integrating maps of current habitat with estimates of prevalence should empower conservation practitioners and decision makers and thus guide actions and policy worldwide.     

Fish assemblage compositions after implementation of the IndVal method on the Narew River system

Indicator species index (IndVal) was used as a new method for an already published study, and allowed of a more convincing way of fish assemblages characterization in a large river system. Three sites clusters (AB, CD, EF) were distinguish using the self-organizing map (SOM, Artificial neural network algorithm) in the lowland Narew River system, which comprised the most characteristic species of the total of 36 present. AB included Pungitius pungitius, Barbatula barbatula, Gasterosteus aculeatus and Gobio gobio (natural and slightly modified sites from small rivers), EF included Leuciscus idus, Perca fluviatilis, Rutilus rutilus, Blicca bjoerkna, Esox lucius, Lota lota and Alburnus alburnus (sites from the main channel and lower courses of biggest tributaries, and CD without characteristic species (containing sites from small and large river ditches impacted by pollution, engineering and both). The IndVal method applied here gives precise and accurate information on fish species habitat preferences.     

Comparing community structure on shells of the abalone Haliotis midae and adjacent rock: implications for biodiversity

This paper concerns the effects on biodiversity of depletion of the South African abalone Haliotis midae, which is a long-lived species with a large corrugated shell that provides a habitat for diverse benthic organisms. We compared community structure on H. midae shells with that on adjacent rock at three sites (Cape Point and Danger Point sites A and B) and at two different times of the year at one of these sites. Shells of H. midae consistently supported communities that were distinctly different from those on rock. In particular, three species of non-geniculate (encrusting) corallines, Titanoderma polycephalum, Mesophyllum engelhartii and Spongites discoideus, were all found either exclusively or predominantly on shells, whereas another non-geniculate coralline, Heydrichia woelkerlingii, occurred almost exclusively on adjacent rock. The primary rocky substratum, however, supported a higher number of species than abalone shells. Possible reasons for the differences between the two substrata include the relative age, microtopography and hardness of the substrata; the abundance of grazers on them; and the relative age of different zones of the abalone shell, which support communities at different stages of succession. Diversity on shells was lowest in zones that were either very young or very old, in keeping with the intermediate disturbance hypothesis. The distinctiveness of shell epibiota will increase β diversity despite having a lower α diversity than that of adjacent rock. Decimation of H. midae by overfishing therefore has implications for biodiversity conservation.     

Modelling the abundance of three key plant species in New Zealand hill-pasture using a decision tree approach

Decision tree models were developed to investigate and predict the relative abundance of three key pasture plants [ryegrass (Lolium perenne), browntop (Agrostis capillaris), and white clover (Trifolium repens)] with integration of a geographical information system (GIS) in a naturalised hill-pasture in the North Island, New Zealand, and were compared with regression models with respect to model fit and predictive accuracy. The results indicated that the decision tree models had a better model fit in terms of average squared error (ASE) and a higher percentage of adequately predicted cases in model validation than the corresponding regression models. These decision tree models clearly revealed the relative importance of environmental and management variables in influencing the abundance of these three species. Hill slope was the most significant environmental factor influencing the abundance of ryegrass while soil Olsen P and annual P fertilizer input were the most significant factors influencing the abundance of browntop, and white clover, respectively. Soil Olsen P of approximately 10 μg/g, or a slope of about 10.5° was critical points where the competition between ryegrass and browntop tended to come to an equilibrium. Integrating the decision tree models with a GIS in this study not only facilitated the model development and analyses, but also provided a useful decision support tool in pasture management such as in assisting precision fertilizer placement. The insights obtained from the decision tree models also have important implications for pasture management, for example, it is important to maintain a soil Olsen P higher than 10 μg/g in order to keep the dominance of ryegrass in the hill-pasture.     

A strong test of a maximum entropy model of trait-based community assembly

We evaluate the predictive power and generality of Shipley's maximum entropy (maxent) model of community assembly in the context of 96 quadrats over a 120-km2 area having a large (79) species pool and strong gradients. Quadrats were sampled in the herbaceous understory of ponderosa pine forests in the Coconino National Forest, Arizona, U.S.A. The maxent model accurately predicted species relative abundances when observed community-weighted mean trait values were used as model constraints. Although only 53% of the variation in observed relative abundances was associated with a combination of 12 environmental variables, the maxent model based only on the environmental variables provided highly significant predictive ability, accounting for 72% of the variation that was possible given these environmental variables. This predictive ability largely surpassed that of nonmetric multidimensional scaling (NMDS) or detrended correspondence analysis (DCA) ordinations. Using cross-validation with 1000 independent runs, the median correlation between observed and predicted relative abundances was 0.560 (the 2.5% and 97.5% quantiles were 0.045 and 0.825). The qualitative predictions of the model were also noteworthy: dominant species were correctly identified in 53% of the quadrats, 83% of rare species were correctly predicted to have a relative abundance of < 0.05, and the median predicted relative abundance of species actually absent from a quadrat was 5 x 10(-5).     

Studies on the family hemiuridae (Trematoda: Digenea), with a description of a new genus from marine fish of Indian waters

A new digenetic trematode, collected 1969–1970 at Ratnagiri (India) from the marine eel Leptocephalus conger, is described as Indostomachicola kinnei n. g., n. sp. The new genus differs from Stomachicola Yamaguti, 1934 and Allostomachicola Yamaguti, 1958 in many characteristics. The new species is characterized by an unlobed ovary, absence of receptaculum seminis, and presence of post-acetabular seminal vesicle and a ventral pit.     

Limitations of Gravity Models in Predicting the Spread of Eurasian Watermilfoil

Abstract: The effects of non‐native invasive species are costly and environmentally damaging, and resources to slow their spread and reduce their effects are scarce. Models that accurately predict where new invasions will occur could guide the efficient allocation of resources to slow colonization. We assessed the accuracy of a model that predicts the probability of colonization of lakes in Wisconsin by Eurasian watermilfoil (Myriophyllum spicatum). We based this predictive model on 9 years (1990–1999) of sequence data of milfoil colonization of lakes larger than 25 ha (n =1803). We used milfoil colonization sequence data from 2000 to 2006 to test whether the model accurately predicted the number of lakes that actually were colonized from among the 200 lakes identified as being most likely to be colonized. We found that a lake's predicted probability of colonization was not correlated with whether a lake actually was colonized. Given the low predictability of colonization of specific lakes, we compared the efficacy of preventing milfoil from leaving occupied sites, which does not require predicting colonization probability, with protecting vacant sites from being colonized, which does require predicting colonization probability. Preventing organisms from leaving colonized sites reduced the likelihood of spread more than protecting vacant sites. Although we focused on the spread of a single species in a particular region, our results show the shortcomings of gravity models in predicting the spread of numerous non‐native species to a variety of locations via a wide range of vectors.     

Application of a Random Forest algorithm to predict spatial distribution of the potential yield of Ruditapes philippinarum in the Venice lagoon, Italy

We present a modelling framework that combines machine learning techniques and Geographic Information Systems to support the management of an important aquaculture species, Manila clam (Ruditapes philippinarum). We use the Venice lagoon (Italy), the first site in Europe for the production of R. philippinarum, to illustrate the potential of this modelling approach. To investigate the relationship between the yield of R. philippinarum and a set of environmental factors, we used a Random Forest (RF) algorithm. The RF model was tuned with a large data set (n = 1698) and validated by an independent data set (n = 841). Overall, the model provided good predictions of site-specific yields and the analysis of marginal effect of predictors showed substantial agreement among the modelled responses and available ecological knowledge for R. philippinarum. The most influent environmental factors for yield estimation were percentage of sand in the sediment, salinity, and water depth. Our results agree with findings from other North Adriatic lagoons. The application of the fitted RF model to continuous maps of all the environmental variables allowed estimates of the potential yield for the whole basin. Such a spatial representation enabled site-specific estimates of yield in different farming areas within the lagoon. We present a possible management application of our model by estimating the potential yield under the current farming distribution and comparing it to a proposed re-organization of the farming areas. Our analysis suggests a reduction of total yield is likely to result from the proposed re-organization.