Maximizing utility of a multispecies, multi-dataset monitoring effort for Arctic raptors in western Alaska 

Arctic climate change precipitates the need for systematic monitoring efforts designed to assess changing population dynamics and ecosystem dysfunction (Knud et al. 2021). Basic demographic information is lacking for many Arctic species and monitoring efforts for birds, including raptors, often receive sporadic funding in Alaska (Knud et al. 2021). Limited conservation and scientific funding necessitates fiscal responsibility including refining methodology to maximize data utility while making efforts to reduce cost (Bottrill et al. 2008). Indicator species, like raptors, can efficiently revel ecosystem disruptions because of their large home ranges and reliance on the underlying function of the trophic web (Sergio et al. 2008). Further, monitoring efforts should aim to build on existing long-term datasets that are required for many important analyses, particular those pertaining to climate change.

​

Raptor surveys that focus on historically occupied territories (or nests) are a common methodology for estimating breeding population sizes (Brown et al. 2013, Johnson et al. 2019) with the additional benefit of assessing breeding success (Bird and Bildstein 2007). Aerial surveys can be an efficient platform from which to observe species that occurs at low densities and in remote areas, like many raptor species (Parker et al. 2011). Aerial surveys allow for much greater spatial coverage that is necessary to generate an adequate sample size to detect abundance trends despite inter-annual variability. Data from aerial surveys can be compiled to calculate occupancy, productivity, and distributions but poorly designed surveys can introduce bias and requires careful consideration (Booms et al. 2010, Johnson et al. 2019, Davis et al. 2022).

​

Annual survival is a primary driver of population dynamics and a critical component of conservation (Monzón and Friedenberg 2018), particularly for long-lived species (Sæther and Bakke 2000). Climate change is likely to impact the survival of birds, but the direction and magnitude of these effects is likely to be species specific (e.g., Robinson et al. 2007). Assessing annual survival requires marking individuals to be tracked over time, and can include color banding (e.g., Barraquand and Nielsen 2021), pit tagging (Gibbons and Andrews 2004), and more recently genotyping feathers can provide a non-invasive means of tracking individuals (Booms et al. 2011a). Further, integrated population models unify count data and demographic data to provide important advantages that including more precise parameter estimates, increased statistical power, and is flexible to sources of uncertainty (reviewed in Schaub and Abadi 2011). Thus, integrated population models can leverage monitoring efforts to clearly reveal population dynamics and help inform management decisions (Millsap et al. 2022).

Data simulations can play a critical role in ensuring survey design and statistical methods provide adequate statistical power to detect real-world population changes (Johnson et al. 2019). Further, simulations allow researchers to subset existing data to reflect potential cost-saving changes to methodology (e.g., reduced survey replicates or number of sites, alternating years, etc) while estimating the statistical power under the proposed methodology. The goal of such simulations should aim to not only detect changes in abundance, but also allow for powerful associations with important environmental variables.   

​

Here we aim to maximize the efficiency of a multi-species dataset consisting of various data types collected annually for Arctic raptors in western Alaska, by constructing integrated population models­­­. These species-specific models will be informed by occupancy and productivity data collected during helicopter surveys (2005 – current), annual adult and juvenile survival from genetically identified feathers (n = 565 individual Gyrfalcons [2011- current] and n = 2100 Golden Eagle feathers with unknown number of individuals [2011-2015]). Due to variable phenologies among the raptor assemblage, our ability to assess to productivity is particularly limited in Rough-legged Hawks, Gyrfalcons, and Golden Eagles because surveys occur before nestlings reach 80% of their fledging age (Steenhof et al. 2017). Thus, we will install camera traps in selected occupied nests to estimate the relationship between nestling age at the time of survey and the probability of fledging, for each species. Lastly, we will simulate real world population changes and calculate the statistical power to detect these changes under current methodologies and various scenarios with cost saving potential. This research will provide needed guidance to ensure population monitoring is sufficient to detect important population changes and is financially responsible to maximize the efficiency of conservation efforts.