Figure 2 - uploaded by Kent P. McFarland
Content may be subject to copyright.
Training gain for each predictor variable alone (black) and the loss in training gain when the variable is removed from the full model (gray). doi:10.1371/journal.pone.0053986.g002
Source publication
Conservation planning and implementation require identifying pertinent habitats and locations where protection and management may improve viability of targeted species. The winter range of Bicknell's Thrush (Catharus bicknelli), a threatened Nearctic-Neotropical migratory songbird, is restricted to the Greater Antilles. We analyzed winter records f...
Context in source publication
Context 1
... regularized training gain for the full model with all presence records was 2.047. From the jackknife test of variable importance (Fig. 2), elevation was the most important predictor variable as measured by the gain produced by a one-variable model, closely followed by winter temperature variables. The two variables that decreased model gain the most when omitted from the full models were land cover and wprecip, suggesting that these two variables contained the most ...
Similar publications
Species distribution models have been used to predict the distribution of invasive species for conservation planning. Understanding spatial transferability of niche predictions is critical to promote species-habitat conservation and forecasting areas vulnerable to invasion. Grain size of predictor variables is an important factor affecting the accu...
Citations
... Although all extant Cuban records of the species outside of its migratory periods derive from Sierra Maestra, the Bicknell's Thrush winter habitat model of McFarland et al. (2013) indicates that a considerable extent of potentially suitable habitat exists in Cuba in areas outside Sierra Maestra. Recent findings that Bicknell's Thrush is a rare and local overwintering species on Puerto Rico (Rimmer et al. 2019), coupled with ongoing and severe habitat loss on its core Hispaniolan wintering range, highlighted the need for systematic, island-wide surveys on Cuba and an assessment of the protected status of occupied areas on the island. ...
... During 2017-2020, our sampling design included both standardized and opportunistic field surveys of forested habitat in southeastern Cuba, which the McFarland et al. (2013) model predicted as the island's region of greatest occupancy by Bicknell's Thrush. Because most areas of predicted high-and medium-probability habitat occur in remote, high-elevation areas with no roads and few trails, we were unable to use a probabilistic sampling approach, instead conducting surveys in tracts of suitable forest habitat that could be reasonably accessed by foot trails. ...
... Most of these were inside three national parks: PN Turquino, PN Bayamesa, and PN Alejandro de Humboldt (hereafter PN Humboldt; Fig. 1). Within each protected area, we sought to survey a variety of forested habitats across an elevational gradient, including forest types and elevations not predicted by the McFarland et al. (2013) model to support habitat for Bicknell's Thrush. ...
To clarify the distribution, relative abundance, and habitat associations of overwintering Bicknell’s Thrush (Catharus bicknelli) in eastern Cuba, we conducted extensive point count and presence-absence surveys during six winters between 1998–2005 and the four winters of 2017–2020, in a variety of forested habitats at varied elevations. During the earlier period, we detected a total of 84 Bicknell’s Thrushes at 54 of 330 discrete sampling points, while in 2017–2020 we detected 33 individual thrushes at 21 of 497 points. We obtained evidence of clustering, with 2–6 individuals registered simultaneously at 24 (32%) of the 75 points with detections in both sampling periods. All thrushes detected in 1998–2005 and 29 (87%) of those detected in 2017–2020 occupied cloud forest habitats at elevations between approximately 1,250 and 1,850 m above sea level (asl) in Sierra Maestra, from Parque Nacional (PN) Pico Turquino in the west to PN Bayamesa in the east. Extensive surveys in lower-elevation forests of Sierra Maestra and other geographic regions of eastern Cuba documented four individual thrushes in non-cloud forest habitat during February of 2019. These were found at adjacent points at an elevation of 650 m asl on the south slope of PN Alejandro de Humboldt; follow-up surveys one year later detected no thrushes in the area. Our survey results in PN Pico Turquino yielded markedly lower numbers of individuals than those detected during early and mid-winter periods from 1998 to 2005. Our results suggest that Bicknell’s Thrush is a rare overwintering species in Cuba, restricted almost entirely to high-elevation cloud forests. These forests are generally well-protected from direct anthropogenic disturbance or degradation. Our findings further suggest that Cuba harbors an overwintering population of Bicknell’s Thrush second only to Hispaniola’s in abundance, and that Cuba is of strategic importance as a refugium for conservation of this globally vulnerable passerine on its restricted wintering range. Keywords Bicknell’s Thrush, Catharus bicknelli, cloud forest, Cuba, overwintering migrant surveys, winter distribution, winter habitat Resumen Estatus invernal del Tordo de Bicknell (Catharus bicknelli) en el oriente de Cuba • Para esclarecer la distribución, abundancia relativa y asociaciones de hábitat del Tordo de Bicknell (Catharus bicknelli), que inverna en el este de Cuba, llevamos a cabo extensos muestreos de presencia–ausencia y puntos de conteo durante seis inviernos entre 1998 y 2005, y durante los cuatro inviernos de 2017 a 2020. Estos muestreos fueron llevados a cabo en una variedad de hábitats boscosos a diferentes elevaciones. Durante el período inicial, detectamos un total de 84 tordos de Bicknell en 54 de los 330 puntos de muestreo discretos, mientras que en 2017–2020 detectamos 33 individuos en 21 de los 497 puntos. Obtuvimos evidencia de agrupamiento, con la presencia simultánea de 2 a 6 individuos registrados en 24 (32%) de los 75 puntos con detecciones durante ambos períodos de muestreo. Todos los tordos detectados en el período de 1998–2005 y 29 (87%) de los detectados en 2017–2020 ocuparon hábitats de bosque nublado a elevaciones entre aproximadamente 1.250 y 1.850 m sobre el nivel del mar (snm) en la Sierra Maestra, desde el Parque Nacional (PN) Pico Turquino al oeste hasta el PN Bayamesa al este. Extensos muestreos en bosques a menor elevación de la Sierra Maestra y otras regiones geográficas del este de Cuba, se documentaron cuatro individuos de esta especie en hábitats diferentes del bosque nublado durante febrero de 2019. Estos se encontraron en puntos adyacentes, a una altura de 650 m snm, en la ladera sur de PN Alejandro de Humboldt. En los muestreos de seguimiento un año después no se detectó ningún tordo en el área. Los resultados de nuestro estudio en el PN Pico Turquino arrojaron un número marcadamente menor de individuos que los detectados durante los períodos de principio y mediados de los inviernos de 1998 a 2005. Nuestros resultados sugieren que el Tordo de Bicknell es una especie rara que inverna en Cuba, restringida casi por completo a los bosques nublados a grandes elevaciones. Por lo general, estos bosques están bien protegidos de las perturbaciones antropogénicas directas o la degradación. Nuestros hallazgos sugieren además que Cuba alberga una población invernal del Tordo de Bicknell, sólo superada por la de La Española en términos de abundancia, y que Cuba tiene una importancia estratégica como refugio para la conservación de este paseriforme globalmente vulnerable en su restringida área de distribución invernal. Palabras clave bosque nublado, Catharus bicknelli, Cuba, distribución invernal, hábitat invernal, muestreos de migrantes invernales, Tordo de Bicknell Résumé Statut de la Grive de Bicknell (Catharus bicknelli) hivernant dans l’est de Cuba • Pour mieux connaître la répartition, l’abondance relative et les habitats de la Grive de Bicknell (Catharus bicknelli) hivernant dans l’est de Cuba, nous avons réalisé de nombreux points de comptage et relevés de présence–absence au cours de six hivers entre 1998 et 2005 et quatre hivers de 2017 à 2020, dans divers habitats forestiers à différentes altitudes. Au cours de la première période, nous avons détecté au total 84 individus sur 54 des 330 points d’échantillonnage distincts, tandis qu’en 2017–2020, nous en avons détecté 33 sur 21 des 497 points. Au cours des deux périodes d’échantillonnage, nous avons observé des regroupements de 2 à 6 individus enregistrés simultanément sur 24 (32 %) des 75 points. L’ensemble des grives détectées en 1998–2005 et 29 (87 %) individus détectés en 2017–2020 étaient présents dans des habitats de forêt humide d’altitude situés environ entre 1 250 et 1 850 m au-dessus du niveau de la mer dans la Sierra Maestra, du Parc national Pico Turquino à l’ouest au Parc national Bayamesa à l’est. Des prospections approfondies réalisées en février 2019 dans les forêts de basse altitude de la Sierra Maestra et d’autres régions de l’est de Cuba ont permis de documenter la présence de quatre grives dans un habitat différent de la forêt humide d’altitude. Elles ont été détectées sur des points adjacents à une altitude de 650 m sur le versant sud du Parc national Alejandro de Humboldt. Des recherches réalisées un an plus tard dans cette zone n’ont pas permis de détecter l’espèce. Les résultats de notre étude dans le Parc national Pico Turquino font état d’effectifs nettement inférieurs à ceux relevés au cours des périodes de début et de milieu d’hiver entre 1998 et 2005. Nos résultats suggèrent que la Grive de Bicknell est une espèce hivernante rare à Cuba, presque entièrement inféodée aux forêts humides de haute altitude. Ces forêts sont généralement bien protégées des dégradations ou des dérangements anthropiques directs. Nos résultats montrent également que Cuba abrite une population hivernante de Grives de Bicknell dont l’abondance se situe juste après celle d’Hispaniola, et que Cuba revêt une importance stratégique en tant que refuge pour la conservation de ce passereau mondialement vulnérable dans son aire d’hivernage restreinte. Mots clés Catharus bicknelli, Cuba, étude des migrateurs hivernants, forêt humide d’altitude, Grive de Bicknell, habitat hivernal, répartition hivernale
... Delineating a species' distribution and optimal habitat is a necessary step in formulating conservation strategies for species management at the habitat or landscape level (Ortega-Huerta and Peterson 2004). In general, species distribution models (SDM) give a measure of a species' occurrence probability in a geographic region and help in identifying a habitat that is critical for target species management (Araujo and Williams 2000;Graham et al. 2004;McFarland et al. 2013). In Manas, an increase in the population of swamp deer (Islam et al. 2022) demands the implementation of specific habitat management strategies, as well as the identification of environmental, geographical, landscape and anthropogenic factors influencing the habitat suitability of the species. ...
... Species distribution models have assisted biodiversity conservation by integrating research into policy and decision-making processes. In addition, conservation planning and execution need the identification of habitat regions where land preservation and management may increase the viability of a single or group of endangered species (McFarland et al. 2013). We give the first prediction map of suitable habitat for swamp deer in Manas National Park as well as the variables influencing their habitat preference. ...
Chilika lagoon is the first Ramsar site in India located along the East Coast. Prediction of the
eutrophication of such ecosystems is a key approach for a sustainable management perspective
as it helps to formulate a management action plan. In the present study a data-driven modeling
approach, an Artificial neural network (ANN) was used to predict eutrophication in the Chilika
lagoon. Back-propagation neural network model was used to relate the major parameters which
influence the eutrophication indicators such as Total nitrogen (TN), Total phosphorus (TP),
Secchi disc depth (SD), dissolved oxygen (DO), Biological oxygen demand (BOD), pH, Water
Temperature (WT), Turbidity (TURB). The model evidenced an acceptable level of prediction
when compared with the results of the field observations. This model's most important
determinant variables were those with a high Random Forest (RF) model permutation relevance
ranking, which reduced the network's structure and led to a more accurate and effective process.
It demonstrated a high agreement between BOD and turbidity. As per the TLI estimation, the
Chilika lagoon was observed to maintain an oligotrophic condition. However, there was a trophic
switchover between the seasons and sectors. The study evidenced that the ANN was able to
predict the indicators with reasonable accuracy which could be proved as a valuable tool for the
Chilika lagoon. This approach can be considered while the formulation of the sustainable
management and conservation action plan for Chilika and other similar aquatic ecosystems
around the globe.
... Delineating a species' distribution and optimal habitat is a necessary step in formulating conservation strategies for species management at the habitat or landscape level (Ortega-Huerta and Peterson 2004). In general, species distribution models (SDM) give a measure of a species' occurrence probability in a geographic region and help in identifying a habitat that is critical for target species management (Araujo and Williams 2000;Graham et al. 2004;McFarland et al. 2013). In Manas, an increase in the population of swamp deer (Islam et al. 2022) demands the implementation of specific habitat management strategies, as well as the identification of environmental, geographical, landscape and anthropogenic factors influencing the habitat suitability of the species. ...
... Species distribution models have assisted biodiversity conservation by integrating research into policy and decision-making processes. In addition, conservation planning and execution need the identification of habitat regions where land preservation and management may increase the viability of a single or group of endangered species (McFarland et al. 2013). We give the first prediction map of suitable habitat for swamp deer in Manas National Park as well as the variables influencing their habitat preference. ...
In this chapter, stacked species distribution models derived from maximum entropy and random forest models are applied on tree species distribution data from Eswatini to estimate and map taxonomic and phylogenetic diversity and endemism using six indices: species richness (SR), taxonomic weighted endemism (WE), corrected taxonomic weighted endemism (CWE), phylogenetic diversity (PD), weighted phylogenetic endemism (WPE) and corrected weighted phylogenetic endemism (CWPE). In addition, hotspots were identified by mapping the 95% percentile of the values from each index.
... Aquatic species' ranges were best predicted by NHDderived stream variables, but for most of these, land use and cover were also definitive. Our results accord with those of previous studies demonstrating that bioclimatic (i.e., associated with temperature and precipitation) and land cover variables are important to structuring species' ecological niches (Elith et al., 2011;McFarland et al., 2013). We also find evidence of anthropogenic impacts on species current ranges, for example, prevalence of developed land cover classes was a negative predictor for some species' habitat suitability and/or probability of occurrence, as has been observed in other SDM projects (Wang et al., 2020). ...
Section 7 of the Endangered Species Act requires the US Environmental Protection Agency (EPA) to consult with the Services (US Fish and Wildlife Service and National Marine Fisheries Service) over potential pesticide impacts to federally listed species. Consultation is complicated by the large number of pesticide products and listed species, as well as by lack of consensus on best practices for conducting co‐occurrence analyses. Previous work demonstrates that probabilistic estimates of species’ ranges and pesticide use patterns improve these analyses. Here we demonstrate that such estimates can be made for suites of sympatric listed species. Focusing on two watersheds, one in in Iowa and the other in Mississippi, we obtained distribution records for 13 species of terrestrial and aquatic listed plants and animals occurring therein. We used maximum entropy modeling and bioclimatic, topographic, hydrographic, and land cover variables to predict species’ ranges at high spatial resolution. We constructed probabilistic spatial models of use areas for two pesticides based on the United States Department of Agriculture Cropland Data Layer and reduced classification errors by incorporating information on the relationships between individual pixels and their neighbors using object‐based images analysis. We then combined species distribution and crop footprint models to derive overall probability of co‐occurrence of listed species and pesticide use. For aquatic species, we also integrated an estimate of downstream residue transport. We report each separate species‐by‐use‐area co‐occurrence estimate, and also combine these modeled co‐occurrence probabilities across species within watersheds to produce an overall metric of potential pesticide exposure risk for these listed species at the watershed level. We propose that the consultation process between EPA and the Services be based on such batched estimation of probabilistic co‐occurrence for multiple listed species at a regional scale. This article is protected by copyright. All rights reserved.
... The tropical and subtropical forests of the Caribbean islands provide important wintering grounds for numerous Neotropical migrant land birds native to Virginia and the eastern United States. An estimated 23 species consistently overwinter in the Greater Antilles and the Bahamas, including Yellow-bellied Sapsucker, Indigo Bunting, White-eyed Vireo, Bicknell's Thrush, Gray Catbird, and numerous New World warblers (Wunderle and Waide 1994, McFarland et al 2013, 2013. Many of these species can spend 6 months or more of the calendar year on their wintering grounds with the non-breeding period presenting a major life history stage in determining survival and fitness (Norris 2005, Norris andMarra 2007). ...
... The tropical and subtropical forests of the Caribbean islands provide important wintering grounds for numerous Neotropical migrant land birds native to Virginia and the eastern United States. An estimated 23 species consistently overwinter in the Greater Antilles and the Bahamas, including Yellow-bellied Sapsucker, Indigo Bunting, White-eyed Vireo, Bicknell's Thrush, Gray Catbird, and numerous New World warblers (Wunderle and Waide 1994, McFarland et al 2013, 2013. Many of these species can spend 6 months or more of the calendar year on their wintering grounds with the non-breeding period presenting a major life history stage in determining survival and fitness (Norris 2005, Norris andMarra 2007). ...
... Finally, our results demonstrate that conservation of the mountainous habitat of Hispaniola would benefit the species by protecting habitat suitable for nesting. Given that other globally rare and vulnerable species use similar habitat, such as Bicknell's Thrush Catharus bicknelli (McFarland et al. 2013) and Hispaniolan Solenodon Solenodon paradoxus (Rodríguez 2011), conservation actions applied in these regions may have multi-species benefits. ...
The Black-capped Petrel or Diablotin Pterodroma hasitata has a fragmented and declining population estimated at c.1,000 breeding pairs. On land, the species nests underground in steep ravines with dense understorey vegetation. The only confirmed breeding sites are located in the mountain ranges of Hispaniola in the Caribbean, where habitat loss and degradation are continuing threats. Other nesting populations may still remain undiscovered but, to locate them, laborious in situ nest searches must be conducted over expansive geographical areas. To focus nest-search efforts more efficiently, we analysed the environmental characteristics of Black-capped Petrel nesting habitat and modeled suitable habitat on Hispaniola using openly available environmental datasets. We used a univariate generalized linear model to compare the habitat characteristics of active Black-capped Petrel nests sites with those of potentially available sites (i.e. random pseudo-absences). Elevation, distance to coast, and the influence of tree cover and density emerged as important environmental variables. We then applied multivariate generalized linear models to these environmental variables that showed a significant relationship with petrel nesting activity. We used the top performing model of habitat suitability model to create maps of predicted suitability for Hispaniola. In addition to areas of known petrel activity, the model identified possible nesting areas for Black-capped Petrels in habitats not previously considered suitable. Based on model results, we estimated the total area of predicted suitable nesting habitat for Black-capped Petrels on Hispaniola and found that forest loss due to hurricanes, forest fires, and encroachment from agriculture had severely decreased availability of predicted suitable habitat between 2000 and 2018.
Resumen
... Humid broadleaf forests in mountainous regions between 500 and 2,500 m of Costa Rica, Panama, Nicaragua, and Honduras are habitat for Golden-winged Warblers , while highland pine-oak forests (1,100-2,400 m) in northern Central America are occupied by Golden-cheeked Warblers (Rappole et al. 2000). Bicknell's Thrush (Catharus bicknelli) also winter in humid montane broadleaf forests in the Greater Antilles (McFarland et al. 2013). While a disproportionate number of declining and threatened species are associated with montane forest, lowland habitats also have a role. ...
... Studies examining the factors driving presence/abundance, as well as comparing habitat quality, reveal a common pattern among several migratory species, an increasing probability of occurrence or abundance in more humid and more mature forests (Johnson et al. 2006a, McFarland et al. 2013, Ruiz-Sanchez et al. 2017, Céspedes and Bayly 2018. For example, increasing canopy height not only results in higher densities of Canada Warblers in the Andes (Céspedes and Bayly 2018) but also a lower rate of extinction from the wetland habitats occupied by Prothonotary Warblers (Bulluck et al. 2019). ...
Of the approximately 716 bird species that breed in North America, 386 (54%) are considered Nearctic-Neotropical migrants by the U.S. Fish and Wildlife Service. In the past 50 yr, scores of these migratory species, including some once considered common, have declined dramatically. Nearctic-Neotropical migrants normally spend 6-8 months in tropical habitats, making the identification, availability, and management of Neotropical habitats critical issues for their conservation. Yet, for most species, complete and nuanced information about their use of tropical habitats and the relative effects of breeding vs. wintering conditions on survival, productivity, and population trends is not available, though many studies point to Neotropical overwintering habitats as being a strong driver of population change. Particularly important for long-distance Nearctic-Neotropical migrants is an understanding of how "carry-over effects" arise and influence population trends when conditions on wintering grounds and tropical stopover areas affect subsequent reproductive performance on breeding grounds. For example, why some species show strong carry-over effects from tropical habitats while others do not is not fully understood. In recent years, many studies have offered insights into these issues by taking advantage of new scientific methods and technological innovations. In this review, we explore threats facing North American breeding birds that migrate to the Neotropics, summarize knowledge of habitat selection and use on the wintering grounds, describe how conditions at one point in the annual cycle may manifest in subsequent seasons or life history stages, and discuss conservation concerns such as climate change and the potential for phenological mismatch.
... Furthermore, it will become increasingly critical to better understand the quality of low elevation habitat, especially regarding breeding success in managed forests, as this habitat may serve to compensate for high elevation habitat losses in the future. Finally, the most limiting factor impacting populations of this species is thought to be habitat loss on the wintering grounds (McFarland et al. 2013, FWS 2017, and while it essential that we continue to manage and conserve habitat on the breeding grounds, it is critical that we also understand and remediate threats to Bicknell's thrush on the wintering grounds. ...
Forest reliant species may be significantly impacted by forest management practices. Understanding these impacts, and whether they are positive or negative, requires a species- specific understanding of habitat use. Bicknell’s thrush (Catharus bicknelli) is a range-restricted habitat specialist occurring in balsam fir (Abies balsamea) dominated montane forests that have been recently disturbed and are undergoing successional growth. While research investigating this species’ habitat use has been conducted throughout much of its breeding range, knowledge of Bicknell’s thrush habitat use in Maine is lacking. Greater understanding of habitat use in Maine would improve the ability of forest managers to promote conservation of habitat for this species of concern. We documented the use of a method for tracking small songbirds in a landscape disadvantageous to using very-high-frequency (VHF) telemetry. Given that the habitat Bicknell’s thrushes occupy is often characterized by rugged terrain and dense forest conditions, efforts to effectively track this species to estimate home-ranges and evaluate habitat use may be confounded. To ameliorate this, we explored the use of a combination tag with a global positioning system (GPS) and VHF component. All things considered, GPS+VHF telemetry was less expensive than VHF telemetry. However, VHF telemetry via triangulation was more accurate than GPS telemetry by 15.09 m. GPS+VHF tags provided greater spatial coverage by collecting data in areas we were otherwise unable to use VHF telemetry effectively. We conclude that GPS+VHF tags offer a feasible alternative to VHF telemetry in densely forested, rugged field conditions. We discuss the potential advantages and disadvantages to VHF- and GPS- based telemetry and make recommendations to researchers interested in employing these methods on small songbirds. We suggest that researchers consider the use of a GPS+VHF tag attached with a weak link leg-loop harness. We also recommend that while researchers should rely on the GPS component of the tag for the majority of their data, we also encourage researchers to continue to track individuals using the VHF tag component when study objectives deem it necessary. We also captured and tracked 24 Bicknell’s thrushes during 2018-2019 in a harvested and non-harvested study area in Maine, USA, and evaluated the influence of forest structure and composition on habitat selection. At the landscape level, Bicknell’s thrushes demonstrated avoidance of tall canopy heights and a large proportion of hardwood tree. At the home-range level within the harvested area, Bicknell’s thrushes selected increasing numbers of small trees (2.54 to 10 cm dbh) and demonstrated a quadratic relationship for selection of canopy height. Similarly, at the home-range level within the non-harvested area, Bicknell’s thrush demonstrated a quadratic relationship for selection of the number of small trees and canopy height. We concluded that Bicknell’s thrushes use lower elevation forest stands in harvested landscapes in Maine. We recommend quantifying forest structure using LiDAR to identify and prioritize stands for use by Bicknell’s thrush.
... Breeding and wintering ranges of Swainson's Thrush and Gray-cheeked Thrush cover most of boreal North America and northern South America, respectively [33,34]. Bicknell's Thrush are restricted to high-altitude boreal forests in eastern North America during the breeding season [32,38] and the Greater Antilles in winter [32,39]. ...
Migration routes vary greatly among small passerine species and populations. It is now possible to determine the routes over great distances and long periods of time with emerging monitoring networks. We tracked individual Swainson’s Thrush ( Catharus ustulatus ), Bicknell’s Thrush ( Catharus bicknelli ) and Gray-cheeked Thrush ( Catharus minimus ) in northeastern Quebec and compared their migration routes and paces across an array of radio-telelemetry stations in North America. Swainson’s Thrush migrated further inland than the other two species. Individuals from all three species slowed their migration pace in the southeastern United States, and Swainson’s Thrush was more likely to stopover than Bicknell’s Thrush. Although individuals were tagged in a small area within or close to their breeding range, the results document the variability of migration routes between species with similar ecological characteristics and provide detailed material to be used for migration studies with broader taxonomic or ecological scope.
... When the correlation between two LiDAR metrics was |0.7| or larger we retained the more informative metric that generated a larger gain value averaged over 40 bootstrap replications after examining scatterplots to assess potential for non-linear relationships. We used MaxEnt's jackknife test of variable importance to evaluate the relative strengths of each predictor variable and identify the best subset for building a final model (McFarland et al., 2013). Our choice of final model consisted of the fewest number of LiDAR metrics with a predictive performance not significantly different than the model with the highest level of performance. ...
The development of metrics derived from LiDAR to quantify structural attributes of forests has contributed to substantial advances in wildlife-habitat modeling. However, further exploration of the numerous metrics available for quantifying canopy complexity could improve models of forest wildlife habitat while simultaneously increasing understanding of wildlife-habitat relationships. We used the full set of metrics available in the LiDAR-processing software FUSION, including several structural metrics that have not previously been used in published habitat models, to identify those that best quantify structural attributes associated with nest site occupancy by the Northern Spotted Owl (NSO; Strix occidentalis caurina). We identified the best subset of predictor variables for building a parsimonious predictive model using an objective selection process of alternative MaxEnt models. The simple metric maximum canopy height was the single best predictor of NSO occupancy, but three rarely used structural metrics included in our final model provided a novel means of describing the distribution of vegetation throughout the canopy height profile. These metrics critically contributed to the model's ability to distinguish small patches of structurally complex suitable habitat within a matrix of structurally simple intermediate-aged forest. Our results indicate the potential value of rarely used LiDAR metrics readily available for objectively quantifying ecologically important but previously inaccessible habitat attributes for arboreal species.
