Body condition is a crucial and indicative measure of an animal’s fitness, reflecting overall foraging success, habitat quality, and balance between energy intake and energetic investment toward growth, maintenance, and reproduction. Recently, drone-based photogrammetry has provided new opportunities to obtain body condition estimates of baleen whales in one, two or three dimensions (1D, 2D, and 3D, respectively) – a single width, a projected dorsal surface area, or a body volume measure, respectively. However, no study to date has yet compared variation among these methods and described how measurement uncertainty scales across these dimensions. This associated uncertainty may affect inference derived from these measurements, which can lead to misinterpretation of data, and lack of comparison across body condition measurements restricts comparison of results between studies. Here we develop a Bayesian statistical model using known-sized calibration objects to predict the length and width measurements of unknown-sized objects (e.g., a whale). We use the fitted model to predict and compare uncertainty associated with 1D, 2D, and 3D photogrammetry-based body condition measurements of blue, humpback, and Antarctic minke whales – three species of baleen whales with a range of body sizes. The model outputs a posterior predictive distribution of body condition measurements and allows for the construction of highest posterior density intervals to define measurement uncertainty. We find that uncertainty does not scale linearly across multi-dimensional measurements, with 2D and 3D uncertainty increasing by a factor of 1.45 and 1.76 compared to 1D, respectively. Each standardized body condition measurement is highly correlated with one another, yet 2D body area index (BAI) accounts for potential variation along the body for each species and was the most precise body condition metric. We hope this study will serve as a guide to help researchers select the most appropriate body condition measurement for their purposes and allow them to incorporate photogrammetric uncertainty associated with these measurements which, in turn, will facilitate comparison of results across studies.
more »
« less
Combining Regional Habitat Selection Models for Large-Scale Prediction: Circumpolar Habitat Selection of Southern Ocean Humpback Whales
Machine learning algorithms are often used to model and predict animal habitat selection—the relationships between animal occurrences and habitat characteristics. For broadly distributed species, habitat selection often varies among populations and regions; thus, it would seem preferable to fit region- or population-specific models of habitat selection for more accurate inference and prediction, rather than fitting large-scale models using pooled data. However, where the aim is to make range-wide predictions, including areas for which there are no existing data or models of habitat selection, how can regional models best be combined? We propose that ensemble approaches commonly used to combine different algorithms for a single region can be reframed, treating regional habitat selection models as the candidate models. By doing so, we can incorporate regional variation when fitting predictive models of animal habitat selection across large ranges. We test this approach using satellite telemetry data from 168 humpback whales across five geographic regions in the Southern Ocean. Using random forests, we fitted a large-scale model relating humpback whale locations, versus background locations, to 10 environmental covariates, and made a circumpolar prediction of humpback whale habitat selection. We also fitted five regional models, the predictions of which we used as input features for four ensemble approaches: an unweighted ensemble, an ensemble weighted by environmental similarity in each cell, stacked generalization, and a hybrid approach wherein the environmental covariates and regional predictions were used as input features in a new model. We tested the predictive performance of these approaches on an independent validation dataset of humpback whale sightings and whaling catches. These multiregional ensemble approaches resulted in models with higher predictive performance than the circumpolar naive model. These approaches can be used to incorporate regional variation in animal habitat selection when fitting range-wide predictive models using machine learning algorithms. This can yield more accurate predictions across regions or populations of animals that may show variation in habitat selection.
more »
« less
- Award ID(s):
- 2026045
- NSF-PAR ID:
- 10320132
- Date Published:
- Journal Name:
- Remote Sensing
- Volume:
- 13
- Issue:
- 11
- ISSN:
- 2072-4292
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
null (Ed.)Predictions from species distribution models (SDMs) are commonly used in support of environmental decision-making to explore potential impacts of climate change on biodiversity. However, because future climates are likely to differ from current climates, there has been ongoing interest in understanding the ability of SDMs to predict species responses under novel conditions (i.e., model transferability). Here, we explore the spatial and environmental limits to extrapolation in SDMs using forest inventory data from 11 model algorithms for 108 tree species across the western United States. Algorithms performed well in predicting occurrence for plots that occurred in the same geographic region in which they were fitted. However, a substantial portion of models performed worse than random when predicting for geographic regions in which algorithms were not fitted. Our results suggest that for transfers in geographic space, no specific algorithm was better than another as there were no significant differences in predictive performance across algorithms. There were significant differences in predictive performance for algorithms transferred in environmental space with GAM performing best. However, the predictive performance of GAM declined steeply with increasing extrapolation in environmental space relative to other algorithms. The results of this study suggest that SDMs may be limited in their ability to predict species ranges beyond the environmental data used for model fitting. When predicting climate-driven range shifts, extrapolation may also not reflect important biotic and abiotic drivers of species ranges, and thus further misrepresent the realized shift in range. Future studies investigating transferability of process based SDMs or relationships between geodiversity and biodiversity may hold promise.more » « less
-
more » « less
Abstract When navigating heterogeneous landscapes, large carnivores must balance trade‐offs between multiple goals, including minimizing energetic expenditure, maintaining access to hunting opportunities and avoiding potential risk from humans. The relative importance of these goals in driving carnivore movement likely changes across temporal scales, but our understanding of these dynamics remains limited.
Here we quantified how drivers of movement and habitat selection changed with temporal grain for two large carnivore species living in human‐dominated landscapes, providing insights into commonalities in carnivore movement strategies across regions.
We used high‐resolution GPS collar data and integrated step selection analyses to model movement and habitat selection for African lions
Panthera leo in Laikipia, Kenya and pumasPuma concolor in the Santa Cruz Mountains of California across eight temporal grains, ranging from 5 min to 12 hr. Analyses considered landscape covariates that are related to energetics, resource acquisition and anthropogenic risk.For both species, topographic slope, which strongly influences energetic expenditure, drove habitat selection and movement patterns over fine temporal grains but was less important at longer temporal grains. In contrast, avoiding anthropogenic risk during the day, when risk was highest, was consistently important across grains, but the degree to which carnivores relaxed this avoidance at night was strongest for longer term movements. Lions and pumas modified their movement behaviour differently in response to anthropogenic features: lions sped up while near humans at fine temporal grains, while pumas slowed down in more developed areas at coarse temporal grains. Finally, pumas experienced a trade‐off between energetically efficient movement and avoiding anthropogenic risk.
Temporal grain is an important methodological consideration in habitat selection analyses, as drivers of both movement and habitat selection changed across temporal grain. Additionally, grain‐dependent patterns can reflect meaningful behavioural processes, including how fitness‐relevant goals influence behaviour over different periods of time. In applying multi‐scale analysis to fine‐resolution data, we showed that two large carnivore species in very different human‐dominated landscapes balanced competing energetic and safety demands in largely similar ways. These commonalities suggest general strategies of landscape use across large carnivore species.
-
The long shadow of woody encroachment: An integrated approach to modeling grassland songbird habitatmore » « less
Abstract Animals must track resources over relatively fine spatial and temporal scales, particularly in disturbance‐mediated systems like grasslands. Grassland birds respond to habitat heterogeneity by dispersing among sites within and between years, yet we know little about how they make post‐dispersal settlement decisions. Many methods exist to quantify the resource selection of mobile taxa, but the habitat data used in these models are frequently not collected at the same location or time that individuals were present. This spatiotemporal misalignment may lead to incorrect interpretations and adverse conservation outcomes, particularly in dynamic systems. To investigate the extent to which spatially and temporally dynamic vegetation conditions and topography drive grassland bird settlement decisions, we integrated multiple data sources from our study site to predict slope, vegetation height, and multiple metrics of vegetation cover at any point in space and time within the temporal and spatial scope of our study. We paired these predictions with avian mark‐resight data for 8 years at the Konza Prairie Biological Station in NE Kansas to evaluate territory selection for Grasshopper Sparrows (
Ammodramus savannarum ), Dickcissels (Spiza americana ), and Eastern Meadowlarks (Sturnella magna ). Each species selected different types and amounts of herbaceous vegetation cover, but all three species preferred relatively flat areas with less than 6% shrub cover and less than 1% tree cover. We evaluated several scenarios of woody vegetation removal and found that, with a targeted approach, the simulated removal of just one isolated tree in the uplands created up to 14 ha of grassland bird habitat. This study supports growing evidence that small amounts of woody encroachment can fragment landscapes, augmenting conservation threats to grassland systems. Conversely, these results demonstrate that drastic increases in bird habitat area could be achieved through relatively efficient management interventions. The results and approaches reported pave the way for more efficient conservation efforts in grasslands and other systems through spatiotemporal alignment of habitat with animal behaviors and simulated impacts of management interventions. -
Understanding how closely related, sympatric species distribute themselves relative to their environment is critical to understanding ecosystem structure and function and predicting effects of environmental variation. The Antarctic Peninsula supports high densities of krill and krill consumers; however, the region is warming rapidly, with unknown consequences. Humpback whales Megaptera novaeangliae and Antarctic minke whales Balaenoptera bonaerensis are the largest krill consumers here, yet key data gaps remain about their distribution, behavior, and interactions and how these will be impacted by changing conditions. Using satellite telemetry and novel spatial point-process modeling techniques, we quantified habitat use of each species relative to dynamic environmental variables and determined overlap in core habitat areas during summer months when sea ice is at a minimum. We found that humpback whales ranged broadly over continental shelf waters, utilizing nearshore bays, while minke whales restricted their movements to sheltered bays and areas where ice is present. This presents a scenario where minke whale core habitat overlaps substantially with the broader home ranges of humpback whales. While there is no indication that prey is limiting in this ecosystem, increased overlap between these species may arise as climate-driven changes that affect the extent, timing, and duration of seasonal sea ice decrease the amount of preferred foraging habitat for minke whales while concurrently increasing it for humpback whales. Our results provide the first quantitative assessment of behaviorally based habitat use and sympatry between these 2 krill consumers and offers insight into the potential effects of a rapidly changing environment on the structure and function of a polar ecosystem.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Journal Article:
- https://doi.org/10.3390/rs13112074
