Habitat Conservation & Wildlife

Contributed Paper
ROOM: HCCC, Room 25A

8:10AM Monarch Patterns of Site Occupancy in Iowa
Rachel A. Vanausdall; Kevin T. Murphy; Stephen J. Dinsmore; Karen E. Kinkead; Paul W. Frese
The Monarch butterfly (Danaus plexippus) is the focus of large-scale habitat restoration efforts because of recent population declines that are in part due to declines in milkweed (Asclepias sp.), which is essential for egg-laying. From 2006-2015 we monitored Monarchs at >400 public and private properties in Iowa. Our broad objective was to link site occupancy and extinction rates to the presence of milkweed and other host plants, site-specific habitat metrics, and landscape context. We used a robust design occupancy model in Program MARK to estimate site occupancy and extinction as well as detection probabilities. Site occupancy varied annually and was negatively influenced by canopy cover (β = -4.87, SE = 0.02). Site occupancy was also negatively influenced by the proportion of agriculture and woodland within a 1 km buffer of the site, although both effects were small. Annual estimates of site occupancy ranged from 0.30 to 1.00 while detection probability ranged from 0.25 to 0.71. Site extinction probabilities also varied annually and were negatively influenced by the amount of milkweed (β = -19.71, SE = 0.09). Furthermore, extinction probability was negatively influenced by the amount of agriculture within 200 m of the site and positively influenced by the amount of grassland within 500 m of the site; both effects were small. Annual extinction probability estimates ranged from <0.01 to 0.36. Our findings confirm the importance of milkweed availability to the Monarch and offer additional insight into how landscape context also plays a role in site use by this species.
8:30AM Implications of Changing Oak and Maple Dynamics for Understanding Stability of Wildlife Habitats in Southeast Ohio
Donald Radcliffe; Stephen Matthews; David Hix
Many recent studies suggest that oak (Quercus) dominated forests in Eastern North America are gradually being replaced by mesic tree genera such as maple (Acer) and beech (Fagus), mainly due to fire suppression over the past century. These changes have the potential to greatly modify wildlife habitat conditions within forest ecosystems and impact many mammal and bird species. We examined the rate of this change over 20-23 years on 120 permanent plots, on both the Marietta and Athens Units of the Wayne National Forest. The sampled forests were originally selected to represent relatively undisturbed stands (at least 94 years old in 2016). We compared these forests’ rates of change across multiple slope aspects and slope positions, based upon an Ecological Land Type classification system. Preliminary results indicate an almost ubiquitous decline in oaks and a corresponding increase in maple and beech; even slope positions thought of as favorable to oak lacked oak regeneration. White oak (Quercus alba) was our dominant species, as it had highest levels of importance value (IV) ((relative basal area + relative density)/2) of any tree species in the initial sampling period (19.7%). It showed relatively large declines in IV on northeast-facing aspects (-7.52%), with relatively small declines on southwest-facing aspects (-0.79%). In contrast, sugar maple (Acer saccharum) increased across both these aspect classes, +6.9% and +7.2% on northeast and southwest-facing aspects, respectively. These pilot results suggest that oaks are not naturally replacing themselves in mature oak forests of Ohio, even in many ecosystem types thought of as favorable to oak. With many forest wildlife species depending on oak ecosystems it is essential that we align our understanding of the spatial and temporal heterogeneity of forest change with wildlife in mind.
8:50AM Mapping the South’s Forests of the Future
Rachel Greene; Kristine Evans; Michael T. Gray; Todd Jones-Farrand; William G. Wathen
Forests in the Southeastern United States hold substantial environmental, economic, and cultural value. Yet up to 23 million acres of forest may be lost in the next 40 years to pressures from changing land use and climate. Southeastern conservation organizations are interested in implementing programs to retain and protect large areas of forest over the next several decades. However, identifying viable areas to target forest conservation on the landscape remains a challenge. We compiled geospatial elements from existing, planned, and prioritized areas for forest protection as well as datasets about threats to forest retention (e.g., urbanization), socioeconomic value (e.g., timber production) of forests, and where reforestation efforts might be most successful. We developed a region-wide index of forest retention in the Southeastern landscape and associated maps for years 2030, 2040, 2050, and 2060. We used gap-analysis to identify areas where future forest protection will support biodiversity benefits. Approximately 12.8 million acres of forestland are currently protected and classified as Very High on our Forest Retention Index (FRI). An additional 187.5 million acres (equivalent to 70% of study area that is currently forested) are highly prioritized, have a low risk of urbanization, and/or a high socio-economic value, and thus classified as High and Moderate-High on FRI. We classified 28.5 million acres (11.4% of study area currently forested) as Very Low due to high urbanization risk and likely loss to sea level rise. Lands in Very Low include some of the most biodiverse areas of the South. This information is presently being used by forest planners to target specific watersheds for forest conservation efforts.
9:10AM Bird Identification: A Bayesian Decision Process
Douglas H. Johnson
Considerable effort has been devoted in recent years to adjusting counts from bird surveys for birds that were in the surveyed area but were not detected, either because they emitted no cues for detection (they were unavailable) or the cues were not noticed by the observer (they were not perceived). What has been largely overlooked is that the counts themselves actually represent decisions made by observers. Those decisions were reached, not only on the basis of information provided by each bird, but by prior knowledge possessed by the observer. Information about the bird could include its appearance, sounds it makes, its behavior, the habitat in which it occurs, etc. Prior knowledge might include the observer’s expectation of which species to expect, considering the location, date, time of day, habitat, and the like. One consequence of this realization is that observers, even with intensive training, likely will differ in their prior knowledge. A second consequence is that, when information about the identity of a bird is scant, the observer’s prior knowledge will dominate the decision about species identification. Here I explore the various kinds of prior information used by observers, how that information influences recordings of birds, and potential consequences of such prior information.
9:30AM Effects of Inundation Probability and Sediment Accumulation on the Connectivity and Structure of the Rainwater Basin, Nebraska
Bram H.F. Verheijen; Dana M. Varner; David A. Haukos
The Rainwater Basin in south-central Nebraska supports a complex network of ~1100 spatially-isolated shallow, recharge, freshwater wetlands. Many species of plants, insects, and amphibians rely on these wetlands for reproduction and survival, with populations connected during dispersal events among wetlands. However, ponded wetlands are not always available to wildlife, because inundation probability and hydroperiod of wetlands depend on patterns in precipitation. Moreover, inundation probability and hydroperiod of many wetlands have sharply decreased since European settlement because agricultural practices have increased rates of sediment accumulation. Physical loss and reduced inundation probabilities of wetlands have increased the distance among ponded wetlands. As a result, wildlife populations of species with limited dispersal capabilities may become isolated and subject to local extinction if distances among wetlands become too large. Unfortunately, it remains unclear how the inundation probability of playa wetlands affects connectivity and structure of the Rainwater Basin, and if wetlands important in maintaining network connectivity are threatened by increased rates of sediment accumulation. We used network models to assess the role of inundation probability on the connectivity and structure of the Rainwater Basin for a range of dispersal distances. We compared the relative importance of each wetland to its risk of disappearing from the landscape by calculating the amount of agriculture surrounding each wetland to assess effects of sediment accumulation on future network connectivity. We found that lack of ponding of several key wetlands during dry years limits long-distance dispersal through the network for species with low dispersal capabilities. Furthermore, we found that several wetlands with low inundation probabilities or a high risk of disappearing from the landscape due to sediment accumulation were important to maintain network connectivity for most dispersal distances. Protecting or restoring these wetlands should therefore be a high priority for conservation.


Contributed Paper
Location: Huntington Convention Center of Cleveland Date: October 11, 2018 Time: 8:10 am - 9:50 am