A tale of four spurs: habitats, distribution, abundance and movements of Oklahomas wintering longspurs and their co-occuring species

Grasslands are the most threatened ecosystem in North America and have been consistently reduced since the settlement of Euro-Americans through the conversion to farmland. This drastic loss in habitat is why grassland birds have suffered worse population declines over the past fifty years compared to other bird guilds. There is growing evidence that survival during the nonbreeding season has a strong influence on the population trends of migratory species. Therefore, understanding winter survival of migratory grassland bird species may be central to their conservation. Chestnut-collared ((Calcarius ornatus): hereafter CCLO) and Thick-billed Longspur ((Rhynchophanes mccownii): hereafter TBLO) are extreme examples of recent population reduction with an 89% and 86% population reduction from 1966-2017. Though prior studies and first-hand expert knowledge exists about longspur wintering ecology, there remain some major research gaps. In Oklahoma and the entire Great Plains there are fundamental unknowns concerning the habitat needs, mortality risks, flocking behavior, and seasonal movements of at-risk longspur species. This dissertation focuses on answering several baseline questions about the wintering ecology of longspurs and their co-occurring grassland bird species. These questions include: What are the distributions of wintering grassland birds in Oklahoma and the Texas Panhandle (i.e., Southern Great Plains)? How do the densities of the most common grassland birds differ across different sites? How do those densities change throughout and between winters? What landscape level variables affect distributions and predict species richness and diversity. To answer these questions, I use multiple techniques including surveying for all grassland bird species using distance sampling walking transects spread out across 14 field sites in Oklahoma and the Texas Panhandle. Using points collected with distance sampling I was able to map the distributions of potentially suitable habitat. I also sought to determine: What fine-scale habitats longspurs use? How those habitats change across different study sites? What management actions could be beneficial to longspurs to improve their wintering habitat? To answer these questions, I used the longspur point locations to sample the horizontal, vertical, and species composition characteristics of the vegetation that longspurs were actively using and compared those characteristics to the vegetation characteristics that was available. Additionally, I sought to address: How do individual CCLO use space throughout the winter? What are the sex and age ratios that CCLO have in Southwestern Oklahoma? For this I captured and tracked individual CCLO with VHF radio-transmitters at the Wichita Mountains Wildlife Refuge. The first Chapter aims to quantify the recent disparity and to explain why there is one in the number of publications between the seasonality of avian research with regards to breeding and non-breeding. I analyzed publication disparity over the last decade (2010-2020), and expanded on the quantification of seasonal bias using two approaches: 1) a traditional literature review of the avian research published in multiple North American ornithological journals, and 2) various text analyses to automate and explore the literature. I found that the disparity between studies of the breeding and non-breeding grounds has not changed within the last decade, and that the publication of studies on breeding continues to far outpace that of studies on non-breeding. Wintering research emphasized more rudimentary and less involved topics such as species distributions, compared to breeding research, which often included predation rates and nest success. I also proposed various reasons for this bias and proposed potential solutions for overcoming those biases. Chapter 2 addresses the need to understand the distribution and densities of grassland birds within the Southern Great Plains. I used distance sampling walking transects across 14 different sites for three winters (2018-19, 2019-20, 2020-21), throughout the three major grassland ecotypes (shortgrass, mixed-grass, and tallgrass prairies) in Oklahoma and Texas Panhandle. I used this data to determine the density and distribution of grassland birds across space and time. I detected 27,166 individuals of 69 species of grassland bird over the three years of the study. I found that many grassland birds exhibit both inter- and intra-seasonal variance in their densities at various sites. At landscape levels most species were negatively associated with shrubs and positively associated with percent of grassland on the landscape, and several species of longspur, including CCLO and TBLO, were positively associated with wider diurnal temperature ranges. My findings can improve management and conservation of critical grassland areas within the Southern Great Plains and will increase the amount of important non-breeding habitat for many declining grassland bird species. For Chapter 3, I quantified the fine-scale habitat characteristics that the four species of longspur use during winter. The Southern Great Plains is an excellent region for studying the habitat differences between the species as the non-breeding ranges of all four longspur species overlap in this region. My study encompassed large, representative tracts of the three major types of prairie ecosystems (i.e., shortgrass, mixed-grass, and tallgrass prairies) that intersect within the Southern Great Plains. I evaluated the relationship between wintering longspur occupancy and the fine-scale habitat characteristics using a combination of standardized bird surveys and paired occupied-random vegetation plot sampling. Using both randomization tests and classification trees, I characterized longspur habitats and compared associations across the three prairie ecosystems. Habitat use varied among all species, departed at very fine scales, and species’ associations shifted across grassland types. These findings can be applied to the management of grasslands to help develop full-life cycle conservation plans for CCLO and TBLO. The final Chapter focuses on CCLO, as they are the most rapidly declining species of grassland bird in North America, experiencing an 89% population decline during the last five decades. I determined the spatial scale at which management would need to occur to properly manage for the species. I captured and tracked individuals during the winters of 2018-19 and 2019-20 at the Wichita Mountains Wildlife Refuge and calculated their home ranges as both Minimum Convex Polygons (MCP) and Fixed Kernel Density Estimates (KD). Across the two years I captured and banded 116 CCLO (roughly 75% male) and fitted 90 of these with VHF radio transmitters. The winter (December 5 through March 8) home ranges defined by MCP averaged 128.8 ha, while the 95% KD indicated a mean of 29.87 ha. Wintering CCLO used larger areas and displayed higher nomadism than what has been reported for other co-occurring grassland bird species. Therefore, management for this species scales beyond the relatively small areas that longspurs aggregate into flocks within, and will require landscape-level coordination to maintain habitat adequate for effective winter population