AIMS for wildlife: Developing an automated interactive monitoring system to integrate real-time movement and environmental data for true adaptive management.

To effectively manage species and habitats at multiple scales, population and land managers require rapid information on wildlife use of managed areas and responses to landscape conditions and management actions. GPS tracking studies of wildlife are particularly informative to species ecology, habitat use, and conservation. Combining GPS data with administrative data and a diverse suite of remotely sensed, geo-referenced environmental (e.g., climatic) data, would more comprehensively inform how animals interact with and utilize habitats and ecosystems and our goal was to create a conceptual model for a system that would accomplish this - the 'Automated Interactive Monitoring System (AIMS) for Wildlife'. Our objective for this study was to develop a Customized Wildlife Report (CWR) - the first AIMS for Wildlife deliverable product. CWRs collate and summarize our 8-year GPS tracking dataset of ∼11 million locations from 1338 individual (16 species) avifauna and make actionable, real-time data on animal movements and trends in a specific area of interest available to managers and stakeholders for rapid application in day-to-day management. The CWR exemplar presented in this paper was developed to address needs identified by habitat managers of Sacramento National Wildlife Refuge and illustrates the highly specific, information offered and how it contributes to assessing the efficacy of conservation actions while allowing for near real-time adaptive management. The report can be easily customized for any of the thousands of wildlife refuges or regional areas of interest in the United States, emphasizing the broad application of an animal movement data stream. Utilizing diverse, extensive telemetry data streams through scientific collaboration can aid managers and conservation stakeholders with short and long-term research and conservation planning and help address a cadre of issues from local-scale habitat management to improving the understanding of landscape level impacts like drought, wildfire, and climate change on wildlife populations.

Machine learned daily life history classification using low frequency tracking data and automated modelling pipelines: application to North American waterfowl.

BACKGROUND: Identifying animal behaviors, life history states, and movement patterns is a prerequisite for many animal behavior analyses and effective management of wildlife and habitats. Most approaches classify short-term movement patterns with high frequency location or accelerometry data. However, patterns reflecting life history across longer time scales can have greater relevance to species biology or management needs, especially when available in near real-time. Given limitations in collecting and using such data to accurately classify complex behaviors in the long-term, we used hourly GPS data from 5 waterfowl species to produce daily activity classifications with machine-learned models using "automated modelling pipelines". METHODS: Automated pipelines are computer-generated code that complete many tasks including feature engineering, multi-framework model development, training, validation, and hyperparameter tuning to produce daily classifications from eight activity patterns reflecting waterfowl life history or movement states. We developed several input features for modeling grouped into three broad categories, hereafter "feature sets": GPS locations, habitat information, and movement history. Each feature set used different data sources or data collected across different time intervals to develop the "features" (independent variables) used in models. RESULTS: Automated modelling pipelines rapidly developed easily reproducible data preprocessing and analysis steps, identification and optimization of the best performing model and provided outputs for interpreting feature importance. Unequal expression of life history states caused unbalanced classes, so we evaluated feature set importance using a weighted F1-score to balance model recall and precision among individual classes. Although the best model using the least restrictive feature set (only 24 hourly relocations in a day) produced effective classifications (weighted F1 = 0.887), models using all feature sets performed substantially better (weighted F1 = 0.95), particularly for rarer but demographically more impactful life history states (i.e., nesting). CONCLUSIONS: Automated pipelines generated models producing highly accurate classifications of complex daily activity patterns using relatively low frequency GPS and incorporating more classes than previous GPS studies. Near real-time classification is possible which is ideal for time-sensitive needs such as identifying reproduction. Including habitat and longer sequences of spatial information produced more accurate classifications but incurred slight delays in processing.

Migration stopover ecology of Cinnamon Teal in western North America.

Identifying migration routes and fall stopover sites of Cinnamon Teal (Spatula cyanoptera septentrionalium) can provide a spatial guide to management and conservation efforts, and address vulnerabilities in wetland networks that support migratory waterbirds. Using high spatiotemporal resolution GPS-GSM transmitters, we analyzed 61 fall migration tracks across western North America during our three-year study (2017-2019). We marked Cinnamon Teal primarily during spring/summer in important breeding and molting regions across seven states (California, Oregon, Washington, Idaho, Utah, Colorado, and Nevada). We assessed fall migration routes and timing, detected 186 fall stopover sites, and identified specific North American ecoregions where sites were located. We classified underlying land cover for each stopover site and measured habitat selection for 12 land cover types within each ecoregion. Cinnamon Teal selected a variety of flooded habitats including natural, riparian, tidal, and managed wetlands; wet agriculture (including irrigation ditches, flooded fields, and stock ponds); wastewater sites; and golf and urban ponds. Wet agriculture was the most used habitat type (29.8% of stopover locations), and over 72% of stopover locations were on private land. Relatively scarce habitats such as wastewater ponds, tidal marsh, and golf and urban ponds were highly selected in specific ecoregions. In contrast, dry non-habitat across all ecoregions, and dry agriculture in the Cold Deserts and Mediterranean California ecoregions, was consistently avoided. Resources used by Cinnamon Teal often reflected wetland availability across the west and emphasize their adaptability to dynamic resource conditions in arid landscapes. Our results provide much needed information on spatial and temporal resource use by Cinnamon Teal during migration and indicate important wetland habitats for migrating waterfowl in the western United States.