No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Using verified species distribution models to inform the conservation of a rare marine species
Abstract
Given the paucity of data on the distribution of habitats and species for most marine species, particularly those that are rare and in need of protection, there is a need to model species distributions. Using the fan mussel, Atrina fragilis (Pennant 1777), as our case species, the aim of the study was to predict new areas of occurrence for A. fragilis, estimate the extent of potentially suitable areas, determine the proportion of these areas that are included in the recently designated nature conservation MPAs off the west of Scotland and identify possible environmental drivers in the distribution of A. fragilis. West coast of Scotland, UK. Using a point process framework, we modelled presence-only data, including historic records. A quadrat survey employing digital still photography was conducted in areas of high and low suitability to verify the model, and subsequently, a targeted survey was undertaken in areas predicted as highly suitable by the models using towed video cameras. Five environmental variables were of prime importance in explaining the distribution of A. fragilis. The results from the verification survey support model performance. Atrina fragilis was found in 80% of the targeted transects undertaken. Approximately 14% of the total area predicted as suitable for A. fragilis occurred within recently designated marine protected areas (MPAs) indicating considerable potential for recolonization given suitable protection. The verified model suggests that limited presence-only and historical records of rare species can perform well within a SDM framework allowing the identification of further suitable areas. The prominence of bathymetric ruggedness in the models is unexpected, given the understood ecological niche for A. fragilis and pinnids in general, but is consistent with the fact that seabed topography can offer protection from fishing pressure. These results will inform restoration objectives of the MPA network.
... The inclusion of fishing as another explanatory variable is not always possible because of data restrictions, but when available they allow to predict alternative scenarios with no fishing , which can be used to reconstruct baselines (Downie et al., 2021b). Unfortunately, for species whose distribution is heavily affected by trawling, this approach could be biased by the range modification already generated by the pressure (Stirling et al., 2016;Downie et al., 2021a). In this work, we combined for the first time the use of distribution models with the use of past presences to delineate habitat loss for a biogenic habitat (the I. elongata coral gardens). ...
... Terrain variables and BPI are probably some of the most common variables used in the field of distribution models for benthic species (e.g. Stirling et al., 2016;Basalo et al., 2019;de la Torriente et al., 2018;Downie et al., 2021b), including previous works on the distribution of I. elongata which also used terrain variables (Lauria et al., 2017). The main reason for the recurrent use of these variables is that despite they are not directly linked to the biology of the species (as also occurs for depth) they are very good proxies (and quite often the only one available in a GIS format) to other relevant variables. ...
... The authors attributed these differences to the impact of trawling on the distribution of F. quadrangularis in the North Sea (and thus its effects on the relationship between the environmental drivers and the probability of presence). In a similar way, Stirling et al. (2016) found unexpected positive effect of ruggedness on the suitable habitat of the fan mussel, Atrina fragilis linking it with the negative impact of trawling on its distribution. To the best of our knowledge, until now, the impact of trawling on sensitive species using a distribution model has been assessed and mapped by replacing trawling effort with zero following the same method used in our first approach Downie et al., 2021a). ...
The bamboo coral Isidella elongata is an engineering species that forms a characteristic biogenic habitat in the bathyal mud of the Mediterranean Sea. This habitat has been severely reduced in recent decades due to trawling impacts, and there is a growing concern about its conservation status. In this work, the habitat loss of I. elongata was computed using a novel approach that combines the realized niche of the species with the estimation of its past distribution (before trawling) to delineate potential areas of habitat loss with different levels of uncertainty. The realized niche of the species was modelled using only live colonies and including trawling effort as explanatory variable whereas the past distribution was estimated also using the leftovers of dead colonies as presences. Trawling effort had a statistically significant negative effect on the extent of the realized niche of I. elongata, confirming previous results on the impact of this pressure on its distribution. The novel approach used in this work has allowed us to map for the first time several areas of potential habitat loss for I. elongata in the studied area, opening new opportunities to provide this essential information for future management and restoration actions of vulnerable marine ecosystems worldwide.
... Spearman's rank correlation coefficients and significance tests were then applied to data extracted from the remaining predictors with 1,000 random points (Stirling et al., 2016;Lecours et al., 2017). ...
... The final nine predictors were viewed as pairs plots ( Figure 2) to examine any persistent relationships between variables and 1,000 randomly generated points (Stirling et al., 2016). Significant correlations of greater than 0.7 between predictors were considered unacceptable for MAXENT inclusion. ...
• As an increasingly important resource in ecological research, citizen scientists have proven dynamic and cost-effective in the supply of data for use within habitat suitability models. With predictions critical to the provision of effective conservation measures in cryptic marine species, this study delivers baseline ecological data for the Critically Endangered angelshark (Squatina squatina), exploring: (i) seasonal, sex-differentiated distributions; (ii) environmental distribution predictors; and (iii) examining bias-corrected, imperfect citizen science data for use in coastal habitat suitability models with cryptic species.
• Citizen science presence data, comprising over 60,000 hours of sampling effort, were used alongside carefully selected open-source predictor variables, with maxent generating seasonal male and female habitat suitability models for angelsharks in the Canary Islands. A biased prior method was used, alongside two model validation measures to ensure reliability.
• Citizen science data used within maxent suggest that angelshark habitat suitability is low in coastal areas during warmer months, with fewer occurrences despite a negligible change in sampling effort. The prime importance of bathymetry may indicate the importance of depth for reproductive activity and possible diel vertical migration, whereas aspect may act as a proxy for sheltered habitats away from open ocean. Substrate as a predictor of female habitats in spring and summer could imply that soft sediment is sought for birthing areas, assisting in the identification of areas critical to reproductive activity and thus locations that may benefit from spatial protections.
• Model outputs to inform recovery plan development and ecotourism are identified as plausible safeguards of population recovery, whereas the comparison of biased and bias-corrected models highlights some variance between methodologies, with bias-corrected models producing greater areas of habitat suitability. Accordingly, an adaptive framework is provided for the implementation of citizen science data within the modelling of cryptic coastal species distribution.
... Perhaps the greatest ecosystem impact to our offshore shelf seas over the past 120 years, has been to reduce many offshore shellfish and bivalve reefs to lowcomplexity shell, sand and gravel beds (Thurstan et al., 2013), with the first pass of a demersal trawl or dredge being the most destructive for many habitats such as bivalve reefs (Cook et al., 2013). This has been the case with reduction in populations of horse mussel, oyster, blue mussel and fan mussel beds (Stirling et al., 2016;Solandt, 2003). These were hitherto prominent in coastal and shelf ecosystems, and provided various functions: They provided mechanisms of locking in carbon and filtering seawater of contaminants and excess nutrients, and acted as a hard habitat base for many species for subsequent colonisation and attachment. ...
... Fishing continues to have a negative impact on climate mitigation . Biogenic habitat-forming seabed species, including horse mussels, blue mussel beds, flame shell reefs, Sabellaria worm reefs, native oyster beds, fan mussels, coral, bryozoan and hydroid seabeds have all become rare in offshore waters because of decades of bottom trawling (Stirling et al., 2016;Solandt, 2003;Thurston et al., 2013). These species used to live and accrue in vast areas before the advent industrial bottom trawling. ...
A report indicating fishing pressure using bottom towed fishing gears inside UK Protected Areas in offshore waters. Focusing on activity in those protected areas designated to protect the seabed. We overlay sediment data, and provisional carbon storage data from Luisetti et al., (2019) to illustrate the potential carbon mitigation costs, and loss of carbon storage capacity between 2015 and 2040 from continued trawling over shelf-sediments
... La utilización de imágenes satelitales para describir y/o predecir la estructuración y distribución de poblaciones con bajo número de individuos, especies raras o que habiten regiones de compleja geografía, establece una moderna alternativa a las observaciones in situ (Forney et al., 2012;Stirling et al., 2016), ayudando en la comprensión de los procesos físicos y biológicos que ocurren en el océano (Lara et al., 2010). No obstante, pese a que las SDM pertenecen a un campo de investigación en notorio crecimiento (Derville et al., 2018), debemos ser cautos con las limitaciones de la técnica de investigación y considerar el tipo de datos satelitales disponibles y la resolución existente para el modelamiento de hábitat debido a cobertura de nubes, distancia de la costa, nivel de procesamiento, escalas temporal y espacial, entre otros (Becker et al., 2016). ...
... Comprender los patrones de distribución de las especies mediante modelamiento por sensoramiento remoto (Stirling et al., 2016), ha permitido desarrollar diversos estudios en cetáceos, logrando conocer, por ejemplo, las condiciones ambientales y geográficas propicias para establecer zonas y periodos de crianza (Lindsay et al., 2016), o bien, conocer los patrones de migración, fuertemente asociados a la abundancia de fitoplancton, medido como Chl-a Visser et al., 2011), todos estudios enfocados en la ballena jorobada (Megaptera novaeangliae). ...
Considered the largest animal on earth, the blue whale (Balaenoptera musculus) has been categorized as "endangered." In the southern hemisphere, their distribution is still poorly understood, due they inhabit remote areas, the number of individuals is low and they possess high migratory capacity. Studies on the distribution of cetaceans link habitat selection to patterns of movement and abundance of prey, mainly euphausia (Euphausia spp.) Or krill, but the specific conditions of the water column can act as a proxy for the environment are frequently related.
Chilean Patagonia, one of the largest estuarine systems in the world, where fjords form a complex oceanographic environment, suitable for the generation of dense aggregations of krill and foraging of the blue whale, could be a relevant site for the Population recovery. The western side of the inland sea of Chiloé and the Gulf of Corcovado are identified as feeding and breastfeeding areas for blue whales.
Understand the structure and distribution of the population is crucial for the correct management of species. Given the spatial and temporal difficulties of dating cetacean sightings, using satellite oceanographic measurements is an important predictor, especially considering the new studies that indicate the relationship between the species and the environment, allowing the modeling of habitat selection.
This study aims to develop a predictive model of the area where there is the higher probability of habitat use by these whales, based on the oceanographic conditions of the area of interest, through satellite images. We will evaluate the oceanographic variables recorded on a daily basis, obtaining greater resolution to identify processes and variations at local level may be determining the presence of blue whales, obtained in research cruises conducted in the summer months of 2014 to 2017.
... Historically, validation of Maxent predictions has lacked an independent assessment of model performance (Greaves, Mathieu & Seddon, 2006), such as a novel set of presence locations. Recent studies have found ground validation of Maxent has been a suitable method to determine the accuracy of predictions (Stirling et al., 2016). The need for independent validation is especially important for rare species exhibiting a wider knowledge gap in distribution than more common species (Rebelo & Jones, 2010). ...
... Recent studies have found ground validation of Maxent has been a suitable method to determine the accuracy of predictions (Stirling et al., 2016). Our study supports this conclusion and offers a unique method, incorporating historic museum localities to inform an SDM of pertinent habitat variables and validating the localities before conducting the SDM. ...
Background
Rare or narrowly endemic organisms are difficult to monitor and conserve when their total distribution and habitat preferences are incompletely known. One method employed in determining distributions of these organisms is species distribution modeling (SDM).
Methods
Using two species of narrowly endemic burrowing crayfish species as our study organisms, we sought to ground validate Maxent, a commonly used program to conduct SDMs. We used fine scale (30 m) resolution rasters of pertinent habitat variables collected from historical museum records in 2014. We then ground validated the Maxent model in 2015 by randomly and equally sampling the output from the model.
Results
The Maxent models for both species of crayfish showed positive relationships between predicted relative occurrence rate and crayfish burrow abundance in both a Receiver Operating Characteristic and generalized linear model approach. The ground validation of Maxent led us to new populations and range extensions of both species of crayfish.
Discussion
We conclude that Maxent is a suitable tool for the discovery of new populations of narrowly endemic, rare habitat specialists and our technique may be used for other rare, endemic organisms.
... Fishing gears are rarely towed across areas of high ruggedness due to potential damage to gear. Such rugged areas may provide a refuge for some contemporarily rare benthic species that historically had a wider distribution (Shepherd et al., 2012;Stirling et al., 2016), as well as species inhabiting reefs and harder ground. Locating MPAs in topographically complex areas is, therefore, unlikely to reduce exposure to fishing pressure. ...
... However, where the distribution of species has historically been widespread, but is now restricted to such refugia, this strategy may not allow recovery across their wider range. For example, historic accounts suggest the distribution of fan mussels was more widespread, but is now mostly limited to areas of high ruggedness in Scottish waters (Stirling et al., 2016). Consequently, de facto protection from fishing appears to be an important driver of contemporary A. fragilis distribution. ...
MPAs are expected to improve the conservation status of rare and important habitats and species. However, the reduction in anthropogenic pressure due to such management measures is rarely estimated. Although MPA networks may cover a large proportion of the seabed, designated areas that prohibit damaging fishing activity are often much smaller. This case study compares fishing pressure inside and outside areas covered by Scottish MPAs and MPA management measures, further relating its distribution to bottom ruggedness, which influences benthic communities and their exposure to mobile bottom fishing. While 7% of the study region was found to be within MPA boundaries, only 2.5% of the region was subject to management measures that restrict mobile bottom fishing. Taking historical levels of fishing as a benchmark, management measures have been applied to <0.6% of the swept area of existing mobile bottom fishing activity. This may be explained by the higher average seabed ruggedness within MPA management measures, which was also where some key species of conservation interest were found. These findings suggest that protection has been focussed in areas that already act as natural refugia for sensitive benthic species and lie away from the majority of fishing activity. While the measures do not reduce fishing pressure markedly, they do protect relatively pristine habitats from future fishing impacts. MPA management measures had higher average seabed ruggedness, suggesting a bias in protection for species and habitats within such areas.
... On modélise selon les variables environnementales disponibles, d'après des connaissances ou des hypothèses écologiques, favorables à la persistance de l'espèce (Franklin 2010). Lorsqu'elle repose sur des hypothèses solides, cette approche peut se révéler extrêmement informative (Jimenez-Valverde et al. 2007 ;Rodríguez et al. 2007 ;Devictor et al. 2010 ;Stirling et al. 2016). De même, les capacités d'extrapolation offertes par les modèles de niche peuvent être déterminantes lorsqu'on souhaite caractériser l'écologie ou la distribution d'une espèce pour laquelle on a relativement peu de données ou des données hétérogènes issues de schémas opportunistes Powney & Isaac 2015). ...
... First we have to address the question "Which is the best statistical framework to model my data?" This question has been largely explored (Aguirre- Gutiérrez et al. 2013;Guillera-Arroita et al. 2015 ;Duque-Lazo et al. 2016) and in recent years, point process models (PPMs) have shown their strength as a unifying frame to fit presence-only species distribution models (SDMs) with many advantages in model implementation and interpretation Stirling et al. 2016). Then comes the crucial question "do I have enough data?" (Virgili et al. 2018), who can be translated into the important matter of "trust upon models" and particularly in their specific contexts (Guillera-Arroita et al. . ...
En regardant de près les outils juridiques et autres leviers, pour la conservation de la biodiversité, il semblerait que les invertébrés, et notamment les insectes, soient minoritaires ou absents. Ce constat est d’autant plus paradoxal lorsque l’on sait que 2/3 de la diversité biologique est composée par des insectes. Comment cette diversité essentielle pour le fonctionnement des écosystèmes se retrouve-t-elle dans l’angle mort de la conservation ?La première réponse avancée est le manque d’outils techniques pour étudier ces organismes petits et relativement insaisissables. La rencontre avec les nouvelles méthodes techniques pour la détection et l’étude des insectes est plus que jamais nécessaire. En effet, ces leviers permettront de faciliter l’étude de ces organismes, d’augmenter les connaissances et ainsi de développer une conservation plus adéquate. Nous évoquerons deux approches en particulier : la détection avec des outils moléculaires et l’utilisation de modèles statistiques pour l’exploration de la distribution potentielle des espèces.Mais les connaissances sont également fondées sur la demande sociétale. Et les connaissances alimentent elles-mêmes les outils de protection et de conservation de la biodiversité. À l’échelle des invertébrés, des disparités existent, privilégiant les « grands papillons bleus » aux « petits diptères marrons ». De fait, l’enjeu le plus important pour déverrouiller la conservation des insectes réside dans l’humain et la perception qu’il a de cette biodiversité. À travers une approche de psychologie de la conservation, nous sonderons la perception du grand public sur les insectes. De même, avec une approche de recherche-action-participative, nous tenterons d’engager divers acteurs vers la conservation d’un groupe d’insectes ordinaires : les coléoptères coprophages. Notre volonté est de proposer des moyens pour sensibiliser, éduquer et engager la société dans cet enjeu majeur qu’est la conservation de l’entomofaune.
... The descriptive and predictive power of SDMs has proved particularly useful to understanding the spatial patterns of rare species or species living in ecosystems that are technically challenging to survey (Dunn, Buchanan, Cuthbert, Whittingham, & McGowan, 2015;Engler, Guisan, & Rechsteiner, 2004;Stirling, Boulcott, Scott, & Wright, 2016). Given their wide-ranging behaviour, their rarity and the remote habitats they live in, cetaceans fall in this category (Redfern et al., 2006), with added observational challenges due to the high proportion of time they spend below the surface. ...
Aim
Accurate predictions of cetacean distributions are essential to their conservation but are limited by statistical challenges and a paucity of data. This study aimed at comparing the capacity of various statistical algorithms to deal with biases commonly found in nonsystematic cetacean surveys and to evaluate the potential for citizen science data to improve habitat modelling and predictions. An endangered population of humpback whales (Megaptera novaeangliae) in their breeding ground was used as a case study.
Location
New Caledonia, Oceania.
Methods
Five statistical algorithms were used to model the habitat preferences of humpback whales from 1,360 sightings collected over 14 years of nonsystematic research surveys. Three different background sampling approaches were tested when developing models from 625 crowdsourced sightings to assess methods accounting for citizen science spatial sampling bias. Model evaluation was conducted through cross‐validation and prediction to an independent satellite tracking dataset.
Results
Algorithms differed in complexity of the environmental relationships modelled, ecological interpretability and transferability. While parameter tuning had a great effect on model performances, GLMs generally had low predictive performance, SVMs were particularly hard to interpret, and BRTs had high descriptive power but showed signs of overfitting. MAXENT and especially GAMs provided a valuable complexity trade‐off, accurate predictions and were ecologically intelligible. Models showed that humpback whales favoured cool (22–23°C) and shallow waters (0–100 m deep) in coastal as well as offshore areas. Citizen science models converged with research survey models, specifically when accounting for spatial sampling bias.
Main conclusions
Marine megafauna distribution models present specific challenges that may be addressed through integrative evaluation, independent testing and appropriately tuned statistical algorithms. Specifically, controlling overfitting is a priority when predicting cetacean distributions for large‐scale conservation perspectives. Citizen science data appear to be a powerful tool to describe cetacean habitat.
... We applied Maxent's variable permutation importance and marginal response curve to interpret species-habitat relationships (Searcy & Shaffer, 2016;Stirling, Boulcott, Scott, & Wright, 2016). ...
Modeling and mapping species distributions are vital to biodiversity conservation, but challenging for data-limited species whose localities are poorly recorded. Here we examine the utility of three datasets and species distribution models in conservation of seahorses (Hippocampus spp.), a genus of poorly-recorded marine fishes. We collated occurrences from field data of species sightings (SS), peer-reviewed literature (PRL), and fishers local ecological knowledge (LEK) for five seahorse species in China. We modelled seahorse distributions using different combinations of these datasets. We first compared model performance and predictions between PRL and LEK, and then evaluated the impact of adding LEK and/or PRL to SS. Our results indicated that LEK provided higher-resolution maps than PRL and tended to generate slightly better models. There is more predictive consistency between LEK and PRL on presence-probability maps than on presence/absence maps. Adding LEK and/or PRL to SS improved model performance across species. Our study suggests that integrating LEK (and PRL) and limited SS with species distribution models can usefully inform conservation for poorly-recorded species.
... Despite this, studies on general SDM theory and methodology mostly focus on the terrestrial environment (reviewed in Franklin 2009;Peterson et al. 2011). A minority of papers specifically address distribution modelling methods in the marine environment: presence-only algorithms (Cheung et al., 2008;Ready et al., 2010;Beaugrand et al., 2011), algorithm comparisons (MacLeod et al., 2008Palialexis et al., 2011;Šiaulys & Bučas, 2012), 3D modelling (Bentlage et al., 2013), rare species (Stirling et al., 2016), joint SDMs (Torres et al., 2008), ensemble modelling (Downie et al., 2013), scale effects (Pittman & Brown, 2011;Nyström Sandman et al., 2013), null models (Merckx et al., 2011), model selection , pseudo-absence generation (Huang et al., 2011;Coro et al., 2016) and predictor datasets (Tyberghein et al., 2012;Sbrocco & Barber, 2013). ...
... Studies on general SDM theory and methodology, however, focus mostly on terrestrial environments (reviewed in Franklin, 2009;Peterson et al., 2011). A minority of papers specifically address distribution modelling methods in the marine environment: presence-only algorithms (Beaugrand, Lenoir, Ibañez, & Manté, 2011;Cheung, Lam, & Pauly, 2008;Ready et al., 2010), algorithm comparisons (MacLeod, Mandleberg, Schweder, Bannon, & Pierce, 2008;Palialexis, Georgakarakos, Karakassis, Lika, & Valavanis, 2011;Šiaulys & Bučas, 2012), 3D modelling (Bentlage, Peterson, Barve, & Cartwright, 2013), rare species (Stirling, Boulcott, Scott, & Wright, 2016), joint SDMs (Torres, Read, & Halpin, 2008), ensemble modelling (Downie, von Numers, & Boström, 2013), scale effects (Nyström Sandman, Wikström, Blomqvist, Kautsky, & Isaeus, 2013;Pittman & Brown, 2011), null models (Merckx, Steyaert, Vanreusel, Vincx, & Vanaverbeke, 2011), model selection (Verbruggen et al., 2013), pseudo-absence generation (Coro et al., 2016;Huang, Brooke, & Li, 2011) and predictor datasets (Sbrocco & Barber, 2013;Tyberghein et al., 2012). ...
Ideally, datasets for species distribution modelling (SDM) contain evenly sampled records covering the entire distribution of the species, confirmed absences and auxiliary ecophysiological data allowing informed decisions on relevant predictors. Unfortunately, these criteria are rarely met for marine organisms for which distributions are too often only scantly characterized and absences generally not recorded. Here, we investigate predictor relevance as a function of modelling algorithms and settings for a global dataset of marine species.
... We applied Maxent's variable permutation importance and marginal response curve to interpret species-habitat relationships (Searcy & Shaffer, 2016;Stirling, Boulcott, Scott, & Wright, 2016). ...
Aim
To identify useful sources of species data and appropriate habitat variables for species distribution modelling on rare species, with seahorses as an example, deriving ecological knowledge and spatially explicit maps to advance global seahorse conservation.
Location
The shallow seas.
Methods
We applied a typical species distribution model (SDM), maximum entropy, to examine the utility of (1) two versions of habitat variables (habitat occurrences vs. proximity to habitats) and (2) three sources of species data: quality research‐grade (RG) data, quality‐unknown citizen science (CS) and museum‐collection (MC) data. We used the best combinations of species data and habitat variables to predict distributions and estimate species–habitat relations and threatened status for seahorse species.
Results
We demonstrated that using “proximity to habitats” and integrating all species datasets (RG, CS and MC) derived models with the highest accuracies among all dataset variations. Based on this finding, we derived reliable models for 33 species. Our models suggested that only 0.4% of potential seahorse range was suitable to more than three species together; seahorse biogeographic epicentres were mainly in the Philippines; and proximity to sponges was an important habitat variable. We found that 12 “Data Deficient” species might be threatened based on our predictions according to IUCN criteria.
Main conclusions
We highlight that using proper habitat variables (e.g., proximity to habitats) is critical to determine distributions and key habitats for low‐mobility animals; collating and integrating quality‐unknown occurrences (e.g., CS and MC) with quality research data are meaningful for building SDMs for rare species. We encourage the application of SDMs to estimate area of occupancy for rare organisms to facilitate their conservation status assessment.
... For example, the method has been used to predict future distributions and range shifts for Mediterranean fishes (Albouy et al., 2013) as well as changes in suitable habitat for commercially important fish and invertebrates along the Atlantic coast of North America (Kleisner et al., 2017). ENM can be regarded as an important tool for conserving threatened species, and has proven useful for the assessment and planning of protected areas (Kremen et al., 2008;Stirling et al., 2016). Thus, the scattered information on Asian horseshoe crab distributions and their unknown conservation status would benefit from gathering recent occurrences as well as a modeling of suitable habitats in the region. ...
Conservation of horseshoe crabs has recently received increasing attention as several populations are in decline. However, scarce information on their distributions in Southeast Asia is impairing conservation efforts. In this study, we sought to improve our understanding of the geographical range and distinct populations of the three Asian horseshoe crabs species in order to identify optimal conservation areas. We mapped the geographic range of Carcinoscorpius rotundicauda, Tachypleus gigas, and T. tridentatus using recent data from field work, literature, Global Biodiversity Information Facility (GBIF), and unpublished data from our scientific network. The data were correlated with 23 different environmental variables of potential ecological importance for horseshoe crabs using the openModeller webservices, including new tidal variables. Ecological niche models were generated using two algorithms, Maximum Entropy and support vector machine, for the three species under present conditions, and projected into a climate change scenario of 2050. The niches of the Asian horseshoe crabs were mostly determined by tidal regime, chlorophyll A concentrations, depth, distance to land, and sea surface temperature. According to our predictions, horseshoe crabs in Southeast Asia are not expected to experience any severe change in extent and distribution of suitable habitat in the future. In order to conserve Asian horseshoe crabs, we suggest establishing Marine Protected Areas at locations where distinct populations and several species occur, such as northern Vietnam, China, Borneo, and southern Japan.
... In terms of field sampling, we adopted a consistent quadrat-based sampling method commonly employed in field ecology (e.g. Stirling et al. 2016). The choice of quadrat size was based on two main factors: time constraints and the maximum size of dropstones and shells. ...
The ecological significance of ice-rafted dropstones in present-day glacial marine benthic environments has received considerable attention from ecologists, but similar studies based on the geological records of dropstones and associated fossils are rare. In this study, we report statistically significant co-occurrences of ice-rafted dropstones with brachiopod shells in multiple stratigraphic horizons at multiple sites from the Middle Permian Wandrawandian Siltstone of the southern Sydney Basin in southeastern Australia. We analysed the distribution patterns of both dropstones and brachiopod-dominated fossil assemblages by using a quadrat-based sampling method and spatial point pattern analysis. It was revealed that the co-occurrences of ice-rafted dropstones and brachiopod shells are not random; rather, they demonstrate statistically significant and stratigraphically recurring associations that are here interpreted to represent dropstone-associated, brachiopod-dominated palaeoecological communities. In these recurring palaeocommunities, the presence of ice-rafted dropstones is considered to have added habitat complexity and heterogeneity to the benthic environment, especially suited to the settlement of brachiopods. In addition, the sinking of dropstones, from floating ice masses through the water column to the silty seafloor, is interpreted to have enriched the nutrient and oxygen supply to the benthic environment, further aiding the flourishing and maintenance of a diverse and stable marine benthic fauna. © 2018 The Author(s). Published by The Geological Society of London. All rights reserved.
... Currently there are two marine protected areas (MPAs) in the OSPAR network that include A. fragilis as a feature identified for protection; the South-West Deeps (England) and the Small Isles (Scotland) MPAs. The description and details of the early life history of A. fragilis, when coupled with information on the spatial distribution of suitable habitat (Stirling et al., 2016) and sea circulation models, will help inform population-level connectivity estimates for this rare species of conservation concern, and may highlight new areas suitable for designation as MPAs. ...
Measuring dispersal in rare sessile benthic species is important in the development of conservation measures such as MPA networks. However, efforts to understand dispersal dynamics for many species of conservation concern are hampered by a lack of fundamental life-history information. Here we present the first description of larvae of the fan mussel, Atrina fragilis, and examine key life-history traits that affect dispersal. Larval identification was accomplished using complementary molecular and morphologic techniques. Atrina-specific primers were designed by aligning Atrina COI sequences available in GenBank. As none of these were from UK specimens, primers were designed in the most conserved regions found across A. fragilis and its closest relative A. chautardi. A monthly time-series of zooplankton samples (2014-2015) suggests that A. fragilis follows the same pattern in spawning observed for other pinnids at temperate latitudes, with peak spawning in summer and winter. Average shell growth was estimated to be 6 μm d⁻¹ based on presumed daily growth lines on larval shells. Measurements of the larval shell visible through the juvenile shell indicate a length of up to 770 μm at settlement. Using presumed daily growth lines, this translates into a pelagic larval duration of around 4 months.
... (e.g.Fournier, Barbet-Massin, Rome, & Courchamp, 2017;Stirling, Scott, & Wright, 2016). Depth and slope have been cited among the main environmental predictors associated with the zonation of benthic communities on seamounts (De la Torriente et al., 2018; Du Preez, Curtis, & Clarke, 2016; McClain & Lundsten, 2015; Serrano et al., 2017). ...
• An ecologically representative, well‐connected, and effectively managed system of marine protected areas (MPAs) has positive ecological and environmental effects as well as social and economic benefits. Although progress in expanding the coverage of MPAs has been made, the application of management tools has not yet been implemented in most of these areas.
• In this work, distribution models were applied to nine benthic habitats on a Mediterranean seamount within an MPA for conservation purposes. Benthic habitat occurrences were identified from 55 remotely operated vehicle (ROV) transects, at depths from 76 to 700 m, and data derived from multibeam bathymetry. Generalized additive models (GAMs) were applied to link the presence of each benthic habitat to local environmental proxies (depth, slope, backscatter, aspect, and bathymetric position index, BPI).
• The main environmental drivers of habitat distribution were depth, slope, and BPI. Based on this result, five different geomorphological areas were distinguished. A full coverage map indicating the potential benthic habitat distribution on the seamount was obtained to inform spatial management.
• The distribution of those habitats identified as vulnerable marine ecosystems (VMEs) was used to make recommendations on zonation for developing the management plan of the MPA. This process reveals itself as an appropriate methodological approach that can be developed in other areas of the Natura 2000 marine network.
... Niche-based sampling is increasingly used in ecology and conservation studies (Stirling, Boulcott, Scott, & Wright, 2016). We reviewed the 462 studies that referred to Guisan et al., 2006 (ac-cording to Google Scholar in January 2019), and found that only 32 implemented this method with field sampling, of which five compared niche-based sampling to other methods by comparing the number of presences recorded per unit of effort. ...
Sampling efficiency is crucial in order to overcome the data crisis in biodiversity and to understand what drives the distribution of rare species. Adaptive niche‐based sampling (ANBS) is an iterative sampling strategy that relies on the predictions of species distribution models (SDMs). By predicting highly suitable areas to guide prospection, ANBS could improve the efficiency of sampling effort in terms of finding new locations for rare species. Its iterative quality could potentially mitigate the effect of small and initially biased samples on SDMs. In this study, we compared ANBS with random sampling by assessing the gain in terms of new locations found per unit of effort. The comparison was based on both simulations and two field surveys of mountain birds. We found an increasing probability of contacting the species through iterations if the focal species showed specialization in the environmental gradients used for modelling. We also identified a gain when using pseudo‐absences during first iterations, and a general tendency of ANBS to increase the omission rate in the spatial prediction of the species’ niche or habitat. Overall, ANBS is an effective and flexible strategy that can contribute to a better understanding of distribution drivers in rare species.
... Niche-based models are increasingly used in ecology and conservation studies (Stirling et al. 2016); however, of the 462 studies that referred to the NBS method (citing Guisan et al. 2006 __________________________________________________________________________Chapitre 4 161 according to Google Scholar in January 2019), only 32 implemented it with field sampling. ...
Du fait de la croissance démographique et économique de la population humaine, nos sociétés sont de plus en plus dépendantes de la nature, et plus précisément de certains processus biologiques regroupés sous le terme services écosystémiques. Comprendre les mécanismes de réponse des systèmes biologiques face aux changements globaux que nos activités induisent à plusieurs échelles est donc un enjeu scientifique et de société. Cette thèse a été pensée et réalisée en interaction avec plusieurs équipes de recherche et gestionnaires d’espaces naturels dans le cadre du développement d’un suivi temporel des oiseaux de montagne comme indicateurs des conséquences des changements globaux sur la biodiversité. Les massifs montagneux de France sont en effet exposés aux changements de climat et à l’évolution de l’élevage des grands herbivores, une activité exposée aux fluctuations des prix des matières premières et des aides publiques. Les forts gradients bioclimatiques « en facettes » (selon l’exposition) qui caractérisent des massifs montagneux font de ces espaces des modèles d’étude particulièrement intéressants pour étudier les effets des changements globaux sur la biodiversité, mais aussi fortement contingents. L’objet général de ma thèse est d’identifier et de quantifier le rôle respectif de la température, de la structure de la végétation, et des activités d’élevage, sur cette communauté d’oiseaux, afin de mieux comprendre quelles pourraient être les conséquences de changements majeurs de climat et d’usage des terres. Les gradients altitudinaux ont été historiquement étudiés de façon isolée et dans de grandes diversités d’habitats. Suivant les objectifs généraux de la thèse nous avons choisi d’adopter l’approche inverse dans le premier chapitre, en multipliant les sites d’études dans un habitat standardisé (1100 points d’écoute réalisés dans les Alpes et les Pyrénées). Nos résultats montrent que la ressource, la température et la structure de l’habitat influent fortement sur la communauté. De plus 5 des 8 espèces étudiées semblent favorisées par le pâturage. Dans le second chapitre, j’ai testé l’effet de la forte saisonnalité qui caractérise le climat des massifs montagneux tempérés, en testant son effet sur la survie individuelle dans une population de Chocards à bec jaune Pyrrhocorax graculus. Je me suis appuyé pour cela sur un suivi individuel (CMR) d’un millier d’individus mené pendant 30 ans par Anne Delestrade. Les Chocards présentent une survie forte et un patron saisonnier, en interaction avec le sexe des individus, avec une survie plus basse pour les femelles après les hivers et printemps chauds. J’ai ensuite présumé que les passereaux insectivores savent profiter des troupeaux en consommant des insectes coprophages. J’ai mesuré les ratios isotopiques stables de l’azote présents dans les fèces des oiseaux les plus communs pour estimer le niveau trophique de leurs proies, et ainsi tester cette hypothèse et quantifier le mécanisme. On observe dans les deux massifs un décalage très marqué vers le prélèvement d’insectes non-herbivores quand l’intensité de pâturage augmente. Le dernier chapitre évalue le potentiel d’échantillonnages itératifs basés sur des modèles de répartition d’espèces pour augmenter la probabilité de contacter une espèce rare dans de nouvelles localités. Cette étude comprend des simulations et un test de terrain dans les Pyrénées sur la Niverolle alpine Montifringilla nivalis et le Monticole de roche Monticolla saxatilis. Les résultats montrent le fort potentiel de la méthode en pratique, et ses limites, avec une augmentation de la spécificité au détriment d’une augmentation des omissions. En discussion générale, je propose des perspectives de recherche visant à généraliser le lien fort entre régime alimentaire des oiseaux de montagne et troupeaux de mammifères herbivores, et à mieux comprendre la phénologie des populations des pelouses d’altitude en fonction de l’enneigement.
... Increasing numbers of models are being conducted on the distribution of marine species, especially rare ones for which conservation efforts are more urgent (L. M. Robinson et al. 2011;Stirling et al. 2016). Over the last ten years, open source databases on the occurrence of marine species and associated environmental parameters have become available (L. M. Robinson et al. 2011). ...
The whale shark is a globally endangered species that is distributed in tropical and warm temperate waters. This study modeled the present-day and future habitat suitability of this species in the Sunda tectonic plate region in Southeast Asia to identify changes and highlight potential areas for conservation. Presence data from the International Union for Conservation of Nature were combined with six environment variables to model habitat suitability under three climate change scenarios. The present-day model results showed that most areas of high suitability matched the occurrence data. The future models generated revealed a small decrease in habitat suitability on the Sunda plate. In all of the three future models, two areas of high suitability were predicted: the central islands of the Philippines and southern Thailand. Areas of high suitability remained the same in the parts of the Gulf of Thailand, southern Indonesia, and the central islands of the Philippines. Based on this information, suggestions on how to protect the environment in these areas were made, including a regional assessment of the whale shark status and implementing a species recovery plan in Southeast Asia.
... First we have to address the question "Which is the best statistical framework to model my data?" This question has been largely explored (Aguirre-Gutiérrez et al., 2013;Guillera-Arroita et al., 2015;Duque-Lazo et al., 2016) and in recent years, point process models (PPMs) have shown their strength as a unifying framework to fit presence-only species distribution models (SDMs) with many advantages in model implementation and interpretation, which can be obscured in popular software platforms such as MaxEnt (Renner et al., 2015;Stirling et al., 2016). Indeed, easy to use "click-button" platforms such as MaxEnt (Philips et al. 2017) and the Biomod R package (Thuiller et al., 2009) have been described as "black box techniques" because users can ignore the details and nuances of their models and default parameters Ahmed et al., 2015;Philips et al., 2017). ...
Citizen science programs, and particularly atlas schemes based on opportunistic biological records, are very important sources of data for species distribution models and conservation. Nevertheless, these data are prone to bias, particularly when they come from less popular or hard to detect/identify species, such as insects. With such biased data, it is important to evaluate the stability of the model predictions. In recent years, point process models (PPMs) have shown their strength as a unifying framework to fit presence-only species distribution models with many advantages in model implementation and interpretation; PPMs are closely connected to methods already in widespread use in ecology such as MaxEnt and to logistic regression and benefit from being more transparent about resource selection and absence handling. Moreover, there is a well-developed set of tools to fit these models and assess various features of the underlying model, including model stability. However, such tools are currently unavailable when point process models are fitted with a lasso penalty, which has been shown to improve predictive performance. Based on the French citizen science program “Stag beetle Quest”, we propose new methods to assess model stability in this context. The ultimate goal was to develop a set of functions to analyze PPM models with lasso penalties fitted with presence-only data. To assess model stability, we randomly sampled different subsets of locations with varying size from the whole dataset and used the proposed tools to compare fitted intensities and model coefficients. All the developed measures are complementary and can be used to identify at what number of point locations the model stabilizes, which will be dependent on the dataset. Our work presents a new toolbox to explore questions around model stability based on the number of locations in the context of point process models with a lasso penalty and confirms once more the use of the point process modeling framework as a flexible and unifying framework to fit presence-only species distribution models.
... Los MDE son herramientas que han sido de gran ayuda para conocer la distribución de especies con una capacidad de desplazamiento grande, difíciles de ver y en zonas que no son del todo accesibles para los investigadores, en las que se presentan dificultades para realizar estudios permanentes y sistemáticos (Stirling et al., 2016), como es el caso de los cetáceos, grupo en el que además, el número de trabajos son pocos a diferencia de otras especies, en especial de aquellos que cuentan con datos de esfuerzo (Redfern et al., 2006;Derville et al., 2018). Aún existen huecos de información acerca de la distribución estacional de cetáceos en una zona tan peculiar como el AGC, que presenta características oceanográficas diferenciadas del resto del GC que le confieren ser una zona de importancia ecológica, con alta productividad (Lluch-Cota y Arias-Arechiga, 2000; Erisman et al., 2015) e importante para varias especies de cetáceos (Tershy, 1992;Gómez-Gallardo, 2013;Chávez-Andrade, 2006). ...
El Alto Golfo de California (AGC) es una provincia del Golfo de California (GC), la oceanografía y fisiografía de la zona hacen de ésta una región productiva y de alta biodiversidad, incluyendo cetáceos, donde se han registrado ocho especies de misticetos y 23 de odontocetos. Se ha descrito que los patrones de desplazamiento de los cetáceos dependen del movimiento de sus presas, de la reproducción y crianza. Conocer la distribución de las especies es una cuestión básica de ecología y los modelos de distribución de especies (MDE) han permitido tener una alternativa para el manejo y conservación de especies o poblaciones. Sin embargo, la distribución está determinada por diversos factores ecológicos, evolutivos y geográficos, que a su vez hacen que el estudio de la distribución de especies sea un tema complejo. Con el objetivo de estimar la presencias y distribución de los cetáceos en el AGC, en este estudio realizamos MDE para seis especies de cetáceos con requerimientos ecológicos diferentes: Balaenoptera edeni, B. physalus, B. musculus, Delphinus capensis, Tursiops truncatus y Orcinus orca. Para ello, utilizamos Modelos Lineales Generalizados a cinco resoluciones diferentes, con datos de la temporada templada y cálida se relacionaron datos de presencia/ ausencia y frecuencia de avistamientos con ocho variables ambientales. Se realizó la estimación de ausencias verdaderas a partir de las rutas de navegación, que también fueron tomadas como un equivalente del esfuerzo. El mejor modelo para cada especie fue aquel con mayor devianza explicada por las variables ambientales y con menor influencia del esfuerzo. Para todos los modelos de todas las especies y temporadas analizadas, a la resolución de 10 x 10 km se obtuvieron los mejores modelos. En los modelos binomiales de odontocetos, la temperatura superficial del mar (TSS) y el oxígeno molecular disuelto, seguidos por el fitoplancton (PHY) y la productividad primaria (PP), fueron las variables significativas con mayor influencia para la presencia de los delfines. La TSS, el PHY y la PP fueron las variables más importantes para la presencia de misticetos. A excepción de unos pequeños parches, todo el AGC resultó ser una zona favorable para la presencia de los cetáceos. La temperatura superficial del mar y la batimetría (Bat) fueron las variables más significativas en los modelos de frecuencias de todos los cetáceos. Para los misticetos, durante la temporada templada hubo una distribución más importante en las zonas más norteñas del AGC y durante la cálida la zona central y suroeste.. Las estimaciones estuvieron dadas por las necesidades ecológicas y biológicas de cada especie.
... Conversely, in the absence of detailed habitat suitability maps, we assumed that focal species were uniformly present over the whole area of MPAs where they had been identified, an assumption that will result in some degree of overestimation of connectivity and dispersal. For example, subsequent species distribution models for tall sea pen (Greathead et al., 2015) and fan mussel (Stirling et al., 2016) predict much smaller extents than the full pMPA areas, but this information was not available at the time the present bio-physical model was parameterized. These uncertainties highlight the need for more detailed and extensive survey and modelling (e.g. ...
Connectivity is a key consideration in the development of networks of marine protected areas (MPAs). However, little is known about the early life history of many of the epi-benthic animals that these spatial measures try to conserve. Here, a pragmatic approach to consider connectivity in such organisms is adopted, as part of the Scottish nature conservation MPA designation process. The primary tool for the study was a basic bio-physical model, forced by a circulation climatology. In the general absence of comprehensive ecological information, the model accounted for the main biological characteristics of the benthic organisms under consideration of relevance to connectivity, namely, presence, spawning season and pelagic larval duration (PLD). The results showed that some degree of connectivity between MPAs is possible even for species with short PLD although those organisms are more likely to be vulnerable to local pressures, particularly in the case of less widely distributed species and those inhabiting less dispersive inshore locations. For MPAs further offshore and species with longer PLD, our simulations suggested large-scale advection patterns crossing large-scale environmental management boundaries. Although the study was an appropriate contribution to the MPA designation process, further refinements encompassing better basic ecological information, enhanced oceanographic resolution, more realistic representation of biological processes (e.g. spawning, larval behaviour) in the model, species presence within and outside MPAs and substrate suitability maps would provide future useful confidence boundaries around the general patterns derived from our study.
A report by leading fisheries experts shares novel analysis on the scale, context, and impacts of the age-old fishing practice of bottom trawling. The report shares new data and analysis combined with policy recommendations to inspire constructive action around this controversial practice.
In recent years, the use of ecological niche models (ENMs) and species distribution models (SDMs) to explore the patterns and processes behind observed distribution of species has experienced an explosive growth. Although the use of these methods has been less common and more recent in marine ecosystems than in a terrestrial context, they have shown significant increases in use and applications. Herein, we provide a systematic review of 328 articles on marine ENMs and SDMs published between 1990 and 2016, aiming to identify their main applications and the diversity of methodological frameworks in which they are developed, including spatial scale, geographic realm, taxonomic groups assessed, algorithms implemented, and data sources. Of the 328 studies, 48 % were at local scales, with a hotspot of research effort in the North Atlantic Ocean. Most studies were based on correlative approaches and were used to answer ecological or biogeographic questions about mechanisms underlying geographic ranges (64 %). A few attempted to evaluate impacts of climate change (19 %) or to develop strategies for conservation (11 %). Several correlative techniques have been used, but most common was the machine-learning approach Maxent (46 %) and statistical approaches such as generalized additive models GAMs (22 %) and generalized linear models, GLMs (14 %). The groups most studied were fish (23 %), molluscs (16 %), and marine mammals (14 %), the first two with commercial importance and the last important for conservation. We noted a lack of clarity regarding the definitions of ENMs versus SDMs, and a rather consistent failure to differentiate between them. This review exposed a need to know, reduce, and report error and uncertainty associated with species’ occurrence records and environmental data. In addition, particular to marine realms, a third dimension should be incorporated into the modelling process, referring to the vertical position of the species, which will improve the precision and utility of these models. So too is of paramount importance the consideration of temporal and spatial resolution of environmental layers to adequately represent the dynamic nature of marine ecosystems, especially in the case of highly mobile species.
Background. Rare or narrowly endemic organisms are difficult to monitor and conserve when their total distribution and habitat preferences are incompletely known. One method employed in determining distributions of these organisms is species distribution modeling (SDM).
Methods. Using two species of narrowly endemic burrowing crayfish species as our study organisms, we sought to ground validate Maxent, a commonly used program to conduct SDMs. We used fine scale (30 m) resolution rasters of pertinent habitat variables collected from historical museum records in 2014. We then ground validated the Maxent model in 2015 by randomly and equally sampling the output from the model.
Results. The Maxent models for both species of crayfish showed positive relationships between predicted relative occurrence rate and crayfish burrow abundance in both a Receiver Operating Characteristic and generalized linear model approach. The ground validation of Maxent led us to new populations and range extensions of both species of crayfish.
Discussion. We conclude that Maxent is a suitable tool for the discovery of new populations of narrowly endemic, rare habitat specialists and our technique may be used for other rare, endemic organisms.
Background. Rare or narrowly endemic organisms are difficult to monitor and conserve when their total distribution and habitat preferences are incompletely known. One method employed in determining distributions of these organisms is species distribution modeling (SDM).
Methods. Using two species of narrowly endemic burrowing crayfish species as our study organisms, we sought to ground validate Maxent, a commonly used program to conduct SDMs. We used fine scale (30 m) resolution rasters of pertinent habitat variables collected from historical museum records in 2014. We then ground validated the Maxent model in 2015 by randomly and equally sampling the output from the model.
Results. The Maxent models for both species of crayfish showed positive relationships between predicted relative occurrence rate and crayfish burrow abundance in both a Receiver Operating Characteristic and generalized linear model approach. The ground validation of Maxent led us to new populations and range extensions of both species of crayfish.
Discussion. We conclude that Maxent is a suitable tool for the discovery of new populations of narrowly endemic, rare habitat specialists and our technique may be used for other rare, endemic organisms.
The first geographical occurrence record of native forest snails "Orthalicus Beck, 1837" in the State of Santa Catarina/ SC and the Southern Brazil region in general is available (see pp. 33-34) ........
The area under the receiver operating characteristic (ROC) curve, known as the AUC, is currently considered to be the standard method to assess the accuracy of predictive distribution models. It avoids the supposed subjectivity in the threshold selection process, when continuous probability derived scores are converted to a binary presence–absence variable, by summarizing overall model performance over all possible thresholds. In this manuscript we review some of the features of this measure and bring into question its reliability as a comparative measure of accuracy between model results. We do not recommend using AUC for five reasons: (1) it ignores the predicted probability values and the goodness-of-fit of the model; (2) it summarises the test performance over regions of the ROC space in which one would rarely operate; (3) it weights omission and commission errors equally; (4) it does not give information about the spatial distribution of model errors; and, most importantly, (5) the total extent to which models are carried out highly influences the rate of well-predicted absences and the AUC scores.
The aim is to determine the environmental requirements, estimate the extent of suitable habitat for three sea pen species, and assess the implications for marine protected areas (MPAs). The sea pen Funiculina quadrangularis and the habitat associated with two further sea pen species, Virgularia mirabilis and Pennatula phosphorea, are of key conservation importance and are recommended for protection within MPAs. This study models their potential distributions using the MAXimum ENTropy algorithm and assesses these in relation to five possible marine protected areas (pMPAs) proposed for Scottish waters. Metrics relevant to assessing the efficacy of MPAs are also presented. Four environmental variables of prime importance for predicting the presence of all three species of sea pen were identified: mud, minimum salinity, depth, and gravel. The habitat suitability index increased with mud content. The modelled distribution of F. quadrangularis indicated a deeper distribution than V. mirabilis or P. phosphorea and was not present in sediment with gravel content above 30%. Pennatula phosphorea had the smallest area of suitable habitat, while V. mirabilis had the largest. The percentage predicted suitable area for each species that was encompassed by the five pMPAs ranged from 11% for F. quadrangularis to 15% for P. phosphorea. Some of the largest areas predicted as suitable for F. quadrangularis lay outside the pMPAs. The model results indicated differences in the environmental requirements of the three species of sea pen that can be linked to the autecology of each species. Patch sizes, calculated from a binary output of the model, were used to estimate the degree of habitat fragmentation, thereby giving a partial assessment of the adequacy criterion for these pMPAs. The results suggest that potential MPAs within the study area cover sizeable areas of potential sea pen habitat. However, further areas suitable for F. quadrangularis could be considered.
The aim is to determine the environmental requirements, estimate the extent of suitable habitat for three sea pen species, and assess the implications for marine protected areas (MPAs). The sea pen Funiculina quadrangularis and the habitat associated with two further sea pen species, Virgularia mirabilis and Pennatula phosphorea, are of key conservation importance and are recommended for protection within MPAs. This study models their potential distributions using the MAXimum ENTropy algorithm and assesses these in relation to five possible marine protected areas (pMPAs) proposed for Scottish waters. Metrics relevant to assessing the efficacy of MPAs are also presented. Four environmental variables of prime importance for predicting the presence of all three species of sea pen were identified: mud, minimum salinity, depth, and gravel. The habitat suitability index increased with mud content. The modelled distribution of F. quadrangularis indicated a deeper distribution than V. mirabilis or P. phosphorea and was not present in sediment with gravel content above 30%. Pennatula phosphorea had the smallest area of suitable habitat, while V. mirabilis had the largest. The percentage predicted suitable area for each species that was encompassed by the five pMPAs ranged from 11% for F. quadrangularis to 15% for P. phosphorea. Some of the largest areas predicted as suitable for F. quadrangularis lay outside the pMPAs. The model results indicated differences in the environmental requirements of the three species of sea pen that can be linked to the autecology of each species. Patch sizes, calculated from a binary output of the model, were used to estimate the degree of habitat fragmentation, thereby giving a partial
assessment of the adequacy criterion for these pMPAs. The results suggest that potential MPAs within the study area cover sizeable areas of potential sea pen habitat. However, further areas suitable for F. quadrangularis could be considered.
The distribution of the fan mussel Atrina fragilis was studied at two contrasting areas of the north Aegean Sea (Thermaikos Gulf): one routinely trawled and one closed to trawlers for over 25 years. Significant differences were detected between the two areas with decreased values in density and size of A. fragilis individuals at the trawled area. As habitat differences, i.e. sediment composition and bathymetry, had a non-significant effect, extensive trawling activities probably explain the observed results.
Atrina seminuda (Bivalvia: Pinnidae) is distributed naturally along the Atlantic coast of America from North Carolina to the Valdés Peninsula. Ar the San Matías gulf (North Patagonia), its southern distributional limit, it inhabits subtidal areas of fine and coarse sandy bottoms, on depths ranging from 4 to 30 m. The reproductive cycle of a population of A. seminuda was studied from October 1999 to October 2000. Histologic samples of gonad tissue were obtained from adult individuals collected at Piedras Coloradas (40°53′S, 65°04′W). Six stages of gonad development were described on the basis of histo-morphologic characteristics: indifferent, early active, developing, mature, spawning, and spent. This study indicates that A. seminuda exhibits an annual reproductive cycle. Gametogenesis starts at the beginning of autumn and is prolonged during the winter. A high synchronism in maturation and spawning occurs at November and December. The spent stage begins at mid-summer and concludes at the end of that season. Muscle index (MI) fluctuates inversely to the gonad index (GI) along the reproductive season, suggesting that this species uses reserves stored in the adductor muscle for gonadic maturation. A. seminuda is a gonochoric stable species that exhibits, on a 1.05% of the individuals, two types of functional hermaphroditism.
Presence-only data present challenges for selecting thresholds to transform species distribution modeling results into binary outputs. In this article, we compare two recently published threshold selection methods (maxSSS and maxFpb) and examine the effectiveness of the threshold-based prevalence estimation approach. Six virtual species with varying prevalence were simulated within a real landscape in southeastern Australia. Presence-only models were built with DOMAIN, generalized linear model, Maxent, and Random Forest. Thresholds were selected with two methods maxSSS and maxFpb with four presence-only datasets with different ratios of the number of known presences to the number of random points (KP–RPratio). Sensitivity, specificity, true skill statistic, and F measure were used to evaluate the performance of the results. Species prevalence was estimated as the ratio of the number of predicted presences to the total number of points in the evaluation dataset. Thresholds selected with maxFpb varied as the KP–RPratio of the threshold selection datasets changed. Datasets with the KP–RPratio around 1 generally produced better results than scores distant from 1. Results produced by We conclude that maxFpb had specificity too low for very common species using Random Forest and Maxent models. In contrast, maxSSS produced consistent results whichever dataset was used. The estimation of prevalence was almost always biased, and the bias was very large for DOMAIN and Random Forest predictions. We conclude that maxFpb is affected by the KP–RPratio of the threshold selection datasets, but maxSSS is almost unaffected by this ratio. Unbiased estimations of prevalence are difficult to be determined using the threshold-based approach.
Experimental scallop dredging was conducted to assess the vulnerability of emergent epifauna on hard substrates. Three sites
were sampled before and after dredging to examine changes in the coverage of faunal turf (hydroid and bryozoan) assemblages
and the composition of the wider epifaunal community. Each site had an “impact” box that was dredged, a control box that was
in an area that was still open to fishing, and a control box in a special area of conservation (SAC) that had not been fished
for two years. Community composition differed significantly after dredging in two of the three sites, with dredged communities
becoming less similar to those in the SAC. There was no clear evidence that dredging in the impact boxes reduced the coverage
of faunal turfs on hard substrates. However, the coverage of faunal turfs on hard substrates in the SAC was typically greater
than in areas that were still being fished commercially, consistent with a dredging effect. The results highlight the role
that substrate morphology might play in modifying the severity of dredging effects. This has relevance to marine spatial management,
as it suggests that emergent epifauna living on hard substrates that are morphologically suited to dredging, such as pebble
and cobble substrates, could be particularly vulnerable to dredging.
Presence-only data, where information is available concerning species presence but not species absence, are subject to bias due to observers being more likely to visit and record sightings at some locations than others (hereafter "observer bias"). In this paper, we describe and evaluate a model-based approach to accounting for observer bias directly - by modelling presence locations as a function of known observer bias variables (such as accessibility variables) in addition to environmental variables, then conditioning on a common level of bias to make predictions of species occurrence free of such observer bias. We implement this idea using point process models with a LASSO penalty, a new presence-only method related to maximum entropy modelling, that implicitly addresses the "pseudo-absence problem" of where to locate pseudo-absences (and how many). The proposed method of bias-correction is evaluated using systematically collected presence/absence data for 62 plant species endemic to the Blue Mountains near Sydney, Australia. It is shown that modelling and controlling for observer bias significantly improves the accuracy of predictions made using presence-only data, and usually improves predictions as compared to pseudo-absence or "inventory" methods of bias correction based on absences from non-target species. Future research will consider the potential for improving the proposed bias-correction approach by estimating the observer bias simultaneously across multiple species.
Species data held in museum and herbaria, survey data and opportunistically observed data are a substantial information resource. A key challenge in using these data is the uncertainty about where an observation is located. This is important when the data are used for species distribution modelling (SDM), because the coordinates are used to extract the environmental variables and thus, positional error may lead to inaccurate estimation of the species–environment relationship. The magnitude of this effect is related to the level of spatial autocorrelation in the environmental variables. Using local spatial association can be relevant because it can lead to the identification of the specific occurrence records that cause the largest drop in SDM accuracy. Therefore, in this study, we tested whether the SDM predictions are more affected by positional uncertainty originating from locations that have lower local spatial association in their predictors. We performed this experiment for Spain and the Netherlands, using simulated datasets derived from well known species distribution models (SDMs). We used the K statistic to quantify the local spatial association in the predictors at each species occurrence location. A probabilistic approach using Monte Carlo simulations was employed to introduce the error in the species locations. The results revealed that positional uncertainty in species occurrence data at locations with low local spatial association in predictors reduced the prediction accuracy of the SDMs. We propose that local spatial association is a way to identify the species occurrence records that require treatment for positional uncertainty. We also developed and present a tool in the R environment to target observations that are likely to create error in the output from SDMs as a result of positional uncertainty.
Summary Modeling the spatial distribution of a species is a fundamental problem in ecology. A number of modeling methods have been developed, an extremely popular one being MAXENT, a maximum entropy modeling approach. In this article, we show that MAXENT is equivalent to a Poisson regression model and hence is related to a Poisson point process model, differing only in the intercept term, which is scale-dependent in MAXENT. We illustrate a number of improvements to MAXENT that follow from these relations. In particular, a point process model approach facilitates methods for choosing the appropriate spatial resolution, assessing model adequacy, and choosing the LASSO penalty parameter, all currently unavailable to MAXENT. The equivalence result represents a significant step in the unification of the species distribution modeling literature.
The life history characteristics of some elasmobranchs make them particularly vulnerable to fishing mortality; about a third of all species are listed by the IUCN as Threatened or Near Threatened. Marine Protected Areas (MPAs) have been suggested as a tool for conservation of elasmobranchs, but they are likely to be effective only if such populations respond to fishing impacts at spatial-scales corresponding to MPA size. Using the example of the Celtic Sea, we modelled elasmobranch biomass (kg h(-1)) in fisheries-independent survey hauls as a function of environmental variables and 'local' (within 20 km radius) fishing effort (h y(-1)) recorded from Vessel Monitoring Systems data. Model selection using AIC suggested strongest support for linear mixed effects models in which the variables (i) fishing effort, (ii) geographic location and (iii) demersal fish assemblage had approximately equal importance in explaining elasmobranch biomass. In the eastern Celtic Sea, sampling sites that occurred in the lowest 10% of the observed fishing effort range recorded 10 species of elasmobranch including the critically endangered Dipturus spp. The most intensely fished 10% of sites had only three elasmobranch species, with two IUCN listed as Least Concern. Our results suggest that stable spatial heterogeneity in fishing effort creates de facto refugia for elasmobranchs in the Celtic Sea. However, changes in the present fisheries management regime could impair the refuge effect by changing fisher's behaviour and displacing effort into these areas.
One of the objectives of forestry planning is to set out criteria for a territory's reforestation oriented towards the reduction of fragmentation and the conservation of biodiversity. This objective may be attained by establishing for each forest type appropriate suitability models, which express the suitability of each point of the territory for the growth of each forest formation. The suitability models may be constructed by utilizing spatial analysis methods, which relate the current presence/absence of forest type to a set of environmental variables. On the basis of maps of diverse environmental variables, we elaborated suitability models for the forests present in the study area using logistic regression and weights-of -evidence techniques integrated into a geographic information system. Combining the suitability models for each forest type using simple comparison operators allowed us to construct a potential vegetation map to use as an objective orientation to the forestry potential of the territory.
The distributions of the pennatulid anthozoans (sea pens) Virgularia
mirabilis, Pennatula
phosphorea and Funiculina
quadrangularis in Scottish waters have previously only been extensively documented in sea lochs, by the Marine Nature Conservation Review (MNCR) surveys. Funiculina
quadrangularis in particular was thought to be restricted to the deep basins of sea lochs and is classified as a nationally uncommon species. Sea pens are characterizing species in seven biotopes found in Scottish waters and are protected under the ‘Mud in Deep Water Habitat Action Plan’ of the UK Biodiversity Action Plan (UKBAP) and are also recorded in several Special Areas of Conservation. The present study has extended the existing information available on sea pens in Scottish waters. The distribution and abundance of the three sea pen species were mapped using GIS (Geographical Information System) by utilizing data from Nephrops
norvegicus stock assessment video surveys carried out by Fisheries Research Services (FRS) Marine Laboratory. Distribution was shown to be aggregated for all species. Funiculina
quadrangularis was present in areas outside the sea lochs, such as the South Minch and the outer mouths of most sea lochs. This study demonstrates that analysis of video footage derived from Nephrops stock assessment surveys can significantly extend our knowledge of sea pen distribution and population density. Future management of Nephrops stocks can be expected to consider impacts of fishing on non-commercial species such as sea pens as a part of the ‘ecosystem approach’ to management.
Transforming the results of species distribution modelling from probabilities of or suitabilities for species occurrence to presences/absences needs a specific threshold. Even though there are many approaches to determining thresholds, there is no comparative study. In this paper, twelve approaches were compared using two species in Europe and artificial neural networks, and the modelling results were assessed using four indices: sensitivity, specificity, overall prediction success and Cohen's kappa statistic. The results show that prevalence approach, average predicted probability/suitability approach, and three sensitivity-specificity-combined approaches, including sensitivity-specificity sum maximization approach, sensitivity-specificity equality approach and the approach based on the shortest distance to the top-left corner (0,1) in ROC plot, are the good ones. The commonly used kappa maximization approach is not as good as the afore-mentioned ones, and the fixed threshold approach is the worst one. We also recommend using datasets with prevalence of 50% to build models if possible since most optimization criteria might be satisfied or nearly satisfied at the same time, and therefore it's easier to find optimal thresholds in this situation.
The distribution of the fan mussel Atrina fragilis was studied at two contrasting areas of the north Aegean Sea (Thermaikos Gulf): one routinely trawled and one closed to trawlers for over 25 years. Significant differences were detected between the two areas with decreased values in density and size of A. fragilis individuals at the trawled area. As habitat differences, i.e. sediment composition and bathymetry, had non-significant effect, extensive trawling activities probably explain the observed results.
Atrina seminuda (Bivalvia:Pinnidae) is distributed naturally along the Atlantic coast of America from North Carolina to the Valdes Peninsula. At the San Matias gulf (North Patagonia), its southern distributional limit, it inhabits subtidal areas of fine and coarse sandy bottoms, on depths ranging from 4 to 30 m. The reproductive cycle of a population of A. seminuda was studied from October 1999 to October 2000. Histologic samples of gonad tissue were obtained from adult individuals collected at Piedras Coloradas (40degrees53'S, 65degrees04'W). Six stages of gonad development were described on the basis of histo-morphologic characteristics: indifferent, early active, developing, mature, spawning, and spent. This study indicates that A. seminuda exhibits an annual reproductive cycle. Gametogenesis starts at the beginning of autumn and is prolonged during the winter. A high synchronism in maturation and spawning occurs at November and December. The spent stage begins at mid-summer and concludes at the end of that season. Muscle index (MI) fluctuates inversely to the gonad index (GI) along the reproductive season, suggesting that this species uses reserves stored in the adductor muscle for gonadic maturation. A. seminuda is a gonochoric stable species that exhibits, on a 1.05% of the individuals, two types of functional hermaphroditism.
The bivalve Atrina fragilis was collected by means of bottom trawl during 17 cruises of PIPETA Expedition in the Adriatic Sea in the period from 1982 to 1994 on the surface of 59000 km2 of continental shelf. A total of 780 hauls were done. Atrina has been recorded at 20% of bottom trawl hauls. The species is the most abundant on sand-silt-clay and clayey "relict" sand in depths between 25 and 50 m. The deepest occurrence was between 251 and 256 m. In spite of intensive trawl fishing and occasional hypoxia events Atrina is renewing.
Spatial and spatio-temporal data are everywhere. Besides those we collect ourselves (‘is it raining?’), they confront us on television, in newspapers, on route planners, on computer screens, on mobile devices, and on plain paper maps. Making a map that is suited to its purpose and does not distort the underlying data unnecessarily is however not easy. Beyond creating and viewing maps, spatial data analysis is concerned with questions not directly answered by looking at the data themselves. These questions refer to hypothetical processes that generate the observed data. Statistical inference for such spatial processes is often challenging, but is necessary when we try to draw conclusions about questions that interest us.
Spatial statistics have been widely applied in epidemiology to the study of the distribution of disease. As we have already shown in, displaying the spatial variation of the incidence of a disease can help us to detect areas where the disease is particularly prevalent, which may lead to the detection of previously unknown risk factors. As a result of the growing interest, Spatial Epidemiology (Elliott etal.,2000) has been established as a new multidisciplinary area of research in recent years.
1.Presence-only data are widely used for species distribution modelling, and point process regression models are a exible tool that has considerable potential for this problem, when data arise as point events.2.In this paper we review point process models, some of their advantages, and some common methods of fitting them to presence-only data.3.Advantages include (and are not limited to): clarification of what the response variable is that is modelled; a framework for choosing the number and location of quadrature points (commonly referred to as pseudoabsences or \background points”) objectively; clarity of model assumptions and tools for checking them; models to handle spatial dependence between points when it is present; ways forward regarding difficult issues such as accounting for sampling bias.4.Point process models are related to some common approaches to presenceonly species distribution modelling, which means that a variety of different software tools can be used to fit these models, including MAXENT or generalised linear modelling software.This article is protected by copyright. All rights reserved.
Species distribution models (SDMs) are used to inform a range of ecological, biogeographical and conservation applications. However, users often underestimate the strong links between data type, model output and suitability for end-use. We synthesize current knowledge and provide a simple framework that summarizes how interactions between data type and the sampling process (i.e. imperfect detection and sampling bias) determine the quantity that is estimated by a SDM. We then draw upon the published literature and simulations to illustrate and evaluate the information needs of the most common ecological, biogeographical and conservation applications of SDM outputs. We find that, while predictions of models fitted to the most commonly available observational data (presence records) suffice for some applications, others require estimates of occurrence probabilities, which are unattainable without reliable absence records. Our literature review and simulations reveal that, while converting continuous SDM outputs into categories of assumed presence or absence is common practice, it is seldom clearly justified by the application's objective and it usually degrades inference. Matching SDMs to the needs of particular applications is critical to avoid poor scientific inference and management outcomes. This paper aims to help modellers and users assess whether their intended SDM outputs are indeed fit for purpose.
Species distribution models (SDMs) offer great potential for inclusion into the toolbox of today's marine environmental manager, especially with regard to marine conservation and planning. The application of SDMs in the marine environment over recent years has been varied but there are still relatively few examples in comparison with terrestrial application, and this is especially true in deep-sea marine ecosystems. This short article builds upon two recent review articles concerning the application of species distribution modelling studies in the marine realm, offering additional practical considerations for discussion. Recommendations for progressing the improved application of SDMs to support marine conservation planning are given, including combining model outputs with other data layers, metadata standards and model error. SDMs have both an urgent and long term contribution to make to marine conservation planning globally, and it is hoped that this article, in combination with developing research on marine SDMs, will contribute to some much needed discussion and inform best practice and new research to enable these models to be of greater use to marine managers.
Species distribution modeling (SDM) links ecological theory of species–environment relationships with statistical learning methods and geospatial data to understand and predict the distributions of species and their habitats. Also called ecological niche modeling, SDM is widely used for biodiversity assessment and to predict the impacts of environmental change on biodiversity in terrestrial and aquatic habitats. It can also provide insight and understanding about ecological relationships.
From the reviews of the First Edition."An interesting, useful, and well-written book on logistic regression models . . . Hosmer and Lemeshow have used very little mathematics, have presented difficult concepts heuristically and through illustrative examples, and have included references."—Choice"Well written, clearly organized, and comprehensive . . . the authors carefully walk the reader through the estimation of interpretation of coefficients from a wide variety of logistic regression models . . . their careful explication of the quantitative re-expression of coefficients from these various models is excellent."—Contemporary Sociology"An extremely well-written book that will certainly prove an invaluable acquisition to the practicing statistician who finds other literature on analysis of discrete data hard to follow or heavily theoretical."—The StatisticianIn this revised and updated edition of their popular book, David Hosmer and Stanley Lemeshow continue to provide an amazingly accessible introduction to the logistic regression model while incorporating advances of the last decade, including a variety of software packages for the analysis of data sets. Hosmer and Lemeshow extend the discussion from biostatistics and epidemiology to cutting-edge applications in data mining and machine learning, guiding readers step-by-step through the use of modeling techniques for dichotomous data in diverse fields. Ample new topics and expanded discussions of existing material are accompanied by a wealth of real-world examples-with extensive data sets available over the Internet.
AimSpecies distribution models (SDMs) are increasingly used to address numerous questions in ecology, biogeography, conservation biology and evolution. Surprisingly, the crucial step of selecting the most relevant variables has received little attention, despite its direct implications for model transferability and uncertainty. Here, we aim to address this with a continent-wide, evaluation of which climate predictors provided the most accurate SDMs for bird distributions. LocationConterminous United States. Methods
For 243 species, we used yearly data since 1971 (from the North American Breeding Bird Survey) to run SDMs (six different algorithms) with combinations of six relatively uncorrelated climate predictors (selected from 22 widely used climate variables). We then estimated the importance of each predictor - both spatially and over a 40-year time period - by comparing the accuracy of the model obtained with or without a given predictor. ResultsThree temperature-related variables (annual potential evapotranspiration, mean annual temperature and growing degree days) produced significantly more accurate SDMs than any other predictors. Among precipitation predictors, annual precipitation provided the most accurate results. Albeit only rarely used in SDMs, the moisture index performed similarly strongly. Interestingly, predictors that summarize average annual climate produced more accurate distributions than seasonal predictors, despite distinct seasonal movements in most species considered. Encouragingly, spatial and temporal (over 40years) evaluation of variables yielded very similar results. Main conclusionsThe approach presented here allowed us to identify the statistically most relevant predictors for birds in the USA and can be applied to other taxa and/or in different parts of the world. Appropriately selecting the most relevant predictors of species distributions at large spatial scale is vital to identifying ecologically meaningful relationships that provide the most accurate predictions under climate change or biological invasions.
With the rise of new powerful statistical techniques and GIS tools, the development of predictive habitat distribution models has rapidly increased in ecology. Such models are static and probabilistic in nature, since they statistically relate the geographical distribution of species or communities to their present environment. A wide array of models has been developed to cover aspects as diverse as biogeography, conservation biology, climate change research, and habitat or species management. In this paper, we present a review of predictive habitat distribution modeling. The variety of statistical techniques used is growing. Ordinary multiple regression and its generalized form (GLM) are very popular and are often used for modeling species distributions. Other methods include neural networks, ordination and classification methods, Bayesian models, locally weighted approaches (e.g. GAM), environmental envelopes or even combinations of these models. The selection of an appropriate method should not depend solely on statistical considerations. Some models are better suited to reflect theoretical findings on the shape and nature of the species’ response (or realized niche). Conceptual considerations include e.g. the trade-off between optimizing accuracy versus optimizing generality. In the field of static distribution modeling, the latter is mostly related to selecting appropriate predictor variables and to designing an appropriate procedure for model selection. New methods, including threshold-independent measures (e.g. receiver operating characteristic (ROC)-plots) and resampling techniques (e.g. bootstrap, cross-validation) have been introduced in ecology for testing the accuracy of predictive models. The choice of an evaluation measure should be driven primarily by the goals of the study. This may possibly lead to the attribution of different weights to the various types of prediction errors (e.g. omission, commission or confusion). Testing the model in a wider range of situations (in space and time) will permit one to define the range of applications for which the model predictions are suitable. In turn, the qualification of the model depends primarily on the goals of the study that define the qualification criteria and on the usability of the model, rather than on statistics alone.
The area under the receiver operating characteristic (ROC) curve, known as the AUC, is currently considered to be the standard method to assess the accuracy of predictive distribution models. It avoids the supposed subjectivity in the threshold selection process, when continuous probability derived scores are converted to a binary presence-absence variable, by summarizing overall model performance over all possible thresholds. In this manuscript we review some of the features of this measure and bring into question its reliability as a comparative measure of accuracy between model results. We do not recommend using AUC for five reasons: (1) it ignores the predicted probability values and the goodness-of-fit of the model; (2) it summarises the test performance over regions of the ROC space in which one would rarely operate; (3) it weights omission and commission errors equally; (4) it does not give information about the spatial distribution of model errors; and, most importantly, (5) the total extent to which models are carried out highly influences the rate of well-predicted absences and the AUC scores.
To conserve biodiversity, it is necessary to understand how species are distributed and which aspects of the environment determine distributions. In large parts of the world and for the majority of species, data describing distributions are very scarce. Museums, private collections and the historical literature offer a vast source of information on distributions. Records of the occurrence of species from these sources are increasingly being captured in electronic databases and made available over the internet. These records may be very valuable in conservation efforts. However, there are a number of limitations with museum data. These limitations are dealt with in the first part of this review. Even if the limitations of museum data can be overcome, these data present a far-from-complete picture of the distributions of species. Species distribution models offer a means to extrapolate limited information in order to estimate the distributions of species over large areas. The second part of this paper reviews the challenges of developing species distribution models for use with museum data and describes some of the questions that species distribution models have been used to address. Given the rapidly increasing number of museum records of species occurrence available over the internet, a review of their usefulness in conservation and ecology is timely.
A system of grain-size nomenclature of terrigenous sediments and sedimentary rocks is introduced wherein fifteen major textural groups are defined on the ratios of gravel, sand, silt, and clay. Further subdivision of each class is based on the median diameter of each size fraction present. Next, the mineral composition of terrigenous sedimentary rocks is considered. A triangular diagram is used to define eight rock types (orthoquartzite, arkose, graywacke, and five transitional types) based on the mineralogy of the silt-sand-gravel fraction and ignoring clay content. The writer contends that the current practice of calling all clayey sandstones "graywackes" is not valid, inasmuch as it represents a confusion of texture with composition. It is suggested that sedimentary rocks may be best defined by the use of a tripartite name, based on the following pattern-(grain size): (textural maturity) (mineral composition).
1. Species distribution modelling is used increasingly in both applied and theoretical research to predict how species are distributed and to understand attributes of species' environmental requirements. In species distribution modelling, various statistical methods are used that combine species occurrence data with environmental spatial data layers to predict the suitability of any site for that species. While the number of data sharing initiatives involving species' occurrences in the scientific community has increased dramatically over the past few years, various data quality and methodological concerns related to using these data for species distribution modelling have not been addressed adequately. 2. We evaluated how uncertainty in georeferences and associated locational error in occurrences influence species distribution modelling using two treatments: (1) a control treatment where models were calibrated with original, accurate data and (2) an error treatment where data were first degraded spatially to simulate locational error. To incorporate error into the coordinates, we moved each coordinate with a random number drawn from the normal distribution with a mean of zero and a standard deviation of 5 km. We evaluated the influence of error on the performance of 10 commonly used distributional modelling techniques applied to 40 species in four distinct geographical regions. 3. Locational error in occurrences reduced model performance in three of these regions; relatively accurate predictions of species distributions were possible for most species, even with degraded occurrences. Two species distribution modelling techniques, boosted regression trees and maximum entropy, were the best performing models in the face of locational errors. The results obtained with boosted regression trees were only slightly degraded by errors in location, and the results obtained with the maximum entropy approach were not affected by such errors. 4. Synthesis and applications . To use the vast array of occurrence data that exists currently for research and management relating to the geographical ranges of species, modellers need to know the influence of locational error on model quality and whether some modelling techniques are particularly robust to error. We show that certain modelling techniques are particularly robust to a moderate level of locational error and that useful predictions of species distributions can be made even when occurrence data include some error.
Question: What are the main drivers for tree species distribution in the Bavarian Alps? What are the species-specific habitat requirements? Are predictions in accordance with expert knowledge?Location: Bavarian Alps (Southern Germany).Methods: To describe tree species–environment relationships, we established species distribution models for the 14 most common tree species of the region. We combined tree species occurrence data from forest inventories and a vegetation database with environmental data from a digital elevation model, climate maps and soil maps. For modelling, we used generalized additive models (GAM) combined with techniques to account for spatial autocorrelation and uneven coverage of environmental gradients. We developed parsimonious models to judge whether statistical models correspond to models based on expert knowledge.Results: Conceptual models were generally in accordance with expectations. Variables based on average temperatures were the most important predictors in most models. Proxies for soil properties such as water and nutrient availability were statistically significant and generally plausible, but appeared largely redundant for model performance. Altitudinal limits of tree species were generally well represented by models. Most species responded differently to summer and January temperatures, indicating that temperature variables are proxies not only for energy balance, but also for frost damage and drought. Although model building benefits considerably from collation with expert knowledge, there are limitations.Conclusions: Meaningful species distribution models can be obtained from noisy data sets covering only a small fraction of species ranges. Models calibrated with such data sets benefit from hypothesis-driven model building rather than strict data-driven model building. Hence, misleading explanations and predictions can be avoided and uncertainties identified. Nevertheless, projections based on climate scenarios can be substantially improved only with models calibrated on a wider data set. Ideally, environmental gradients should cover the whole niche space of a species, or at least include regions with analogous climate.
Rank abundance distributions (RADs) are a description of community structure common to every ecological sample where counts are recorded and are useful for managing and understanding biodiversity. We use RADs to describe patterns of biodiversity in samples with high numbers of unique species. We use a novel statistical method to analyse RADs and demonstrate prediction methods for attributes of biodiversity. The RAD is defined by the total abundance (Ni), species richness (Si) and the vector of relative abundances (nij) and the joint probability distribution of these quantities is modelled. Models were fitted to benthic biological data sampled on the Western Australian coast from depths of 100 to 1500 m and a latitudinal range of 22 to 35oS, using topographic and oceanographic data as covariates. Predictions from fitted models give attributes of biodiversity derived from RADs at a regular grid over the sampled area. The Leeuwin current and Leeuwin undercurrent appear to be key structuring forces for the predicted biodiversity attributes. The predictions show that benthic biodiversity is complex and varies with a number of different covariates. The predictions are unique, as they characterise important aspects of biodiversity and how it varies with large spatial scales. The predictions enable the complete reconstruction of the expected RAD at any point where covariates are available with estimates of uncertainty.