Pheasants 365: A review of seasonal ring-necked pheasant habitat in central North America

Ring-necked pheasants, an economically important game species in North America, are facing population declines in agriculturally dominated areas due to intensification of production systems. There is a significant body of research on pheasant habitat requirements, much of which aims to address declines by guiding management decisions to improve habitat conditions. Pheasants, like many species, have differing resource needs throughout the year based on their life history and seasonal changes to the landscape; however, many studies have only focused on one or two of the three key ecological seasons – breeding, brood-rearing, or overwintering. We conducted a literature review using the Web of Science database to synthesize results from studies investigating pheasant habitat in central North America in a seasonal context in order to provide a year-round perspective on pheasant habitat management. Our results show the importance of grasslands and small grains during nesting and brood-rearing season, while wetlands and food plots become important during the winter months. Grassland structure is a crucial component of nesting habitat, with pheasants selecting for high litter content, high VOR, and tall vegetation for nesting in 90% of studies that included those variables; while during brood-rearing chicks had higher success in grasslands with a higher percentage of forbs and more bare ground. No strong relationship was found with woodlands during any season, positive or negative. Row crops showed mixed results during nesting and brooding, as they are frequently used but not preferred, while during the winter row crops and other agricultural lands such as pasture and hay fields were largely negatively related to pheasant resource use, abundance, and/or survival. This review aims to guide management decisions regarding pheasant habitat by providing a season-specific synthesis of the literature within the Great Plains and Midwestern USA.

Methods
The geographic scope of this review was the Great Plains and Midwest Region of the United States. In recognition of the variability and subjectivity in delineating the boundaries of these regions, we used the Great Plains Region outlined by Lavin et al. (2011) and the Midwest Region definition provided by National Geographic (O’Conner, 2024). Notably, there is an overlap between the two regions, with portions of North Dakota, South Dakota, Nebraska, and Kansas occurring in both regions. We specified the combined extent of these regions as our study area due to their relative similarities in land use, topography, and climate, as well as history of supporting pheasant populations. The temporal extent of the review was from 1970 – 2023.

A systematic literature search was conducted according to the PRISMA guidelines for ecology and evolution (O’Dea et al., 2021) using the Web of Science ‘all databases’ search function. Search terms and Boolean operators used are reported in Table 1. Subsequently, search results were refined to only include peer-reviewed articles originating from the United States. This yielded a total of 174 articles, which were then subjected to manual review by two independent reviewers to determine if they met inclusion criteria. Papers were included if any portion of the study was conducted within the study area described above. Additional inclusion criteria included: (1) the study must appear in a peer-reviewed journal; (2) must present original research on pheasant habitat, resource use, or survival directly related to habitat; (3) must study wild, in-situ pheasant populations; (4) no genetic studies were included; (5) no diet analyses were included; (6) multi-species studies were included only if they reported species-specific results; (7) biogeographic studies were included only if they evaluated habitat or resource use as a driver of changes in range or distribution. Books, reports, theses, dissertations, grey literature, magazines, newsletters, fact sheets, conferences, seminar or symposium proceedings, assessment projects, reviews, guides, and bibliographies were excluded. Studies using a mix of wild and captive or translocated pheasants were excluded if the majority of pheasants were captive or translocated. The selection process was executed independently by two authors, with outcomes cross-checked to ensure consistent application of the inclusion criteria. After this manual selection, 65 papers remained and were analyzed. The abstract, methods, and results of each study were thoroughly reviewed and summarized in Microsoft Excel. After the relevant information was extracted from each paper, it was synthesized in R (R Core Team, 2024) and interpreted for trends, differences, and commonalities.

Land cover types were hierarchically classified in order to standardize the results of each study for quantitative analysis. Grassland includes both public grasslands and Conservation Reserve Program (CRP) fields; woody cover includes shrublands, woodlands, and shelterbelts; wetland includes herbaceous and woody wetlands; linear features include roadsides, fencerows, waterways, and train rights-of-way; and agricultural grassland includes hay fields, alfalfa, and pastures. Wheat and wheat stubble were incorporated into the small grains category. Percentage or abundance of litter and forbs were included as grassland structural components due to their relevance in the literature to microhabitat and nest site selection (Matthews et al., 2012a). For each analysed paper, each land cover category was designated as “positive”, “negative”, “variable”, “no effect found”, or “not included in analysis”; where, positive/negative are defined as having a positive or negative correlation with pheasant abundance, occupancy, habitat selection, survival, or nest/brood success at at least one spatial scale. A land cover category was labelled as ‘no effect found” if it was measured during the study but did not appear in the top model or a relationship was found to be insignificant (p < 0.05). A land cover category was labelled as “variable” if the direction of the relationship changed based on scale, study area, or some other studied variable (i.e. row crops showed a positive relationship with pheasant abundance at the home range scale but a negative relationship at the landscape scale).