Spatial Analysis For Conservation and Sustainability
Conservation
Biodiversity is threatened, and conservation is urgent. The reality of limited resources for conservation requires prioritization among actions, species, and places, and the building of capacity in countries where threats are high.
Understanding human influence on ecosystems and their services is crucial to achieve sustainable development and ensure the conservation of biodiversity. In this context, the human footprint index (HFI) represents the anthropogenic impacts on ecosystems and the natural environment. Our objective was to characterize the HFI in Southern Patagonia (Argentina) across the landscape, qualifying the differences among the main ecological areas and especially the forested landscapes. We also assessed the potential utility of HFI to identify priority conservation areas according to their wilderness quality and potential biodiversity values. We created a HFI map (scores varied from 0 representing high wilderness quality to 1 representing maximum human impact) using variables related to direct (e.g. infrastructure) and indirect (e.g. derived from economic activities) human impacts, including settlements, accessibility, oil industry, and sheep production. HFI varied significantly across the natural landscapes, being lower (0.07 0.11) in remote ecosystems close to the Andes Mountains and higher (0.38 0.40) in southern areas close to the provincial capital city. Forested landscapes presented different impact values, which were directly related to the economical values of the different forest types. We determined that the current protected area network is not equally distributed across the different ecological areas and forest types. Priority conservation areas were also identified using the fragmentation produced by the human impact, the patch size, and the potential biodiversity values. HFI can present high compatibility with other land-use management decision making tools, acting as a complement to the existing tools for conservation planning or management.
Considering their outsized importance as prey for so many species one would assume that patterns of insect abundance and their determinants have been well-studied. On the contrary, insect ecology is poorly understood and documented. Our study sought to gain an understanding of the subgroup of insects that fly, with a particular emphasis on groups that spend part of their life in lakes and streams.
Insect trapping took place in the northern highland lake district. The study area consists of northern mixed forest with one of the highest densities of lakes in the world.
We conducted insect trapping over three years in the forest landscape of northern Wisconsin, near UW-Madison’s Trout Lake Research Station. We trapped insects May-August around five different lakes and identified them in the lab.
Mean insect abundance varied over the course of the summer with peak average insect abundance occurring in late May and early June (Julian Day 145-155) in both years.
There were several patterns that stood out. Flying insects tended to be many times more abundant in nearshore areas compared to interior forests. Different groups of insects showed different patterns. Diptera, including deerflies, midges, and gnats were the most abundant insects overall. As expected, emergent aquatic groups such as midges, mayflies, and dragonflies were more abundant in nearshore areas while beetles and thrips were more abundant in forest interiors. There were also multiple peaks of abundance through the season with large emergence events of midges and mayflies driving much of the pattern. In addition, local canopy cover was negatively correlated with insect abundance.
We observed birds, bats, and fish consuming flying insects. Abundance of these insect predators likely tracks the abundance of their insect prey. In addition, insects perform other ecosystems services such as pollination and nutrient cycling. Understanding the patterns and drivers of insect abundance can help us better understand northern Wisconsin forest ecosystems.
Prioritizing candidate areas to achieve species richness representation is relatively straightforward when distributions are known for many taxa; however, it may be challenging in data-poor regions. One approach is to focus on the distribution of a few charismatic species in areas that overlap with areas with little human influence, and another is to expand protection in the vicinity of existing protected areas. We assessed the effectiveness of these two approaches for protecting the potential distribution of 21 bird species affiliated with the piedmont dry forest in Argentina. We assessed the degree to which current protected areas met the representation target for each bird species. We found that 8% of the piedmont dry forest and 11% of the extent of occurrence of the bird species within piedmont dry forest were protected, indicating a shortfall. Areas with little human influence that overlap with the distribution of charismatic species had a higher number of bird species than areas with high human influence. Areas within the vicinity of protected areas performed similarly to priority areas, but included high human influence areas. We suggest that a prioritization scheme based on areas of charismatic species distribution that overlap with areas of low human influence can function as an effective surrogate for bird species affiliated with the piedmont dry forest in Argentina. Our results have operational implications for conservation planning in those regions of the world where biodiversity data are poor, but where decisions and actions to sustain biodiversity are urgently needed.
Grassland birds have exhibited dramatic and widespread declines since the mid-20th century. Greater prairie chickens (Tympanuchus cupido pinnatus) are considered an umbrella species for grassland conservation and are frequent targets of management, but their responses to land use and management can be quite variable. We used data collected during 2007–2009 and 2014–2015 to investigate effects of land use and grassland management practices on habitat selection and survival rates of greater prairie chickens in central Wisconsin, USA. We examined habitat, nest-site, and brood-rearing site selection by hens and modeled effects of land cover and management on survival rates of hens, nests, and broods. Prairie chickens consistently selected grassland over other cover types, but selection or avoidance of management practices varied among life-history stages. Hen, nest, and brood survival rates were influenced by different land cover types and management practices. At the landscape scale, hens selected areas where brush and trees had been removed during the previous year, which increased hen survival. Hens selected nest sites in hay fields and brood-rearing sites in burned areas, but prescribed fire had a negative influence on hen survival. Brood survival rates were positively associated with grazing and were highest when home ranges contained ≈15%–20% shrub/tree cover. The effects of landscape composition on nest survival were ambiguous. Collectively, our results highlight the importance of evaluating responses to management efforts across a range of life-history stages and suggest that a variety of management practices are likely necessary to provide structurally heterogeneous, high-quality habitat for greater prairie chickens. Brush and tree removal, grazing, hay cultivation, and prescribed fire may be especially beneficial for prairie chickens in central Wisconsin, but trade-offs among life-history stages and the timing of management practices must be considered carefully.
Human activity cause major changes to the planet and biodiversity is declining at an alarming rates. In order to prevent biodiversity loss, conservation actions require to assess current status of biodiversity to better understand and predict future changes, to identify the major drivers of biodiversity patterns, and to map biodiversity patterns. However, monitoring biodiversity over large areas is challenging to do in the field. Remote sensing provides the opportunity to develop indices that are designed for biodiversity assessment, because satellite data are collected systematically across broad scales. Vegetation productivity is one of the important determinants of species richness and density across broad scale. Vegetation indices derived from satellite data are a good proxy for vegetation productivity over broad areas. The Dynamic Habitat Indices (DHIs) summarize the three different aspects of vegetation productivity: cumulative productivity, minimum productivity, and seasonality in the way that it became relevant for biodiversity (Hobi et al., 2017; Radeloff et al., 2019; Razenkova et al., 2020). However, so far the DHIs have only been derived from coarse-resolution satellite imagery, which limits their value for management decisions.
The DHIs based on Landsat 8 in Colorado US, spatial resolution 30-m.The DHIs based on MODIS in Colorado US, spatial resolution 1-km.
Our goal is to develop the DHIs using medium-resolution Landsat imagery for monitoring biodiversity and abundance pattern across the conterminous United States. The main advantage is that imagery with medium resolution provides more detailed information about the spatial patterns of productivity. Our rationale was that the DHIs with higher spatial resolution could capture the difference in vertical structure of vegetation and characterize habitat heterogeneity at much finer scale, especially in complex mountainous terrain and areas with fragmented land cover. Having this crucial information in my hand, will help to understand how species respond to anthropogenic modification of landscapes, which disturb the integrity of landscape pattern. However, the temporal resolution of Landsat is low, and that creates a lot of challenges for the calculation of the DHIs.
We will develop the DHIs for the conterminous United States and test the usefulness of the DHIs for explaining the avian species richness and abundance pattern. Our study covers a wide range of ecoregions, and has diverse climatic zones and topography, resulting in a large number of habitats and large ranges of the DHIs. Moreover, rich datasets for bird richness and abundance are available for the US, particularly the western US. Our research will add more understanding to the importance of higher spatial resolution for characterizing the DHIs metrics and consequently for modeling biodiversity and individual species pattern. Moreover, our work will add more knowledge about drivers of avian diversity across broad spatial extents that can be used to predict how biodiversity patterns will change in the future depending on changes in vegetation productivity.
Given the rapid rate of climatic change occurring during the winter months, particularly in the Northern Hemisphere, researchers are working diligently to assess the potential effects of these changes on biodiversity assessments and conservation planning. A major hindrance to these efforts to date has been a lack of remotely sensed indices that characterize winter conditions for ecological questions across large spatiotemporal scales. David and Likai are addressing this need by leveraging the wealth of available satellite data to derive new indices of snow and frozen ground dynamics. These Winter Habitat Indices (WHIs) capture several biologically important aspects of winter, including overall length, within-season climate variability, and potential subnivium conditions.
(1) Winter Habitat Indices (WHIs) for the contiguous US derived from MODIS data (500-m) by David Gudex-Cross. Calculated annually from 2000/01 – 2018/19. These indices summarize satellite observations of winter conditions in ecologically meaningful ways.
To date, David has developed three WHIS for the contiguous US using MODIS snow observations and temperature data from Daymet at 500-meter spatial resolution: snow season length, snow cover variability, and the frequency of frozen ground without snow days. First, snow season length is the total number of days between the first and last snow in a given year (a measure of overall winter length). Second, snow cover variability captures the frequency with which a pixel is snow covered then not (ablation) or vice versa (new snow) within a given snow season. This index is especially important for species that rely on white coloration during the winter for camouflage, such as snowshoe hares, because it enables the identification of potential mismatch periods (i.e., when an individual is in its white color morph but there is no snow). Third, the frequency of frozen ground without snow days approximates subnivium conditions, with higher frequencies meaning organisms faced harmful freezing conditions without the thermal refugia provided by snow more often in a given year.
Using MODIS snow observations and freeze/thaw data from microwave sensors, Likai previously developed three WHIs globally at a 25-kilometer spatial resolution: duration of frozen ground, frozen ground with snow, and frozen ground without snow. Now he has also derived the snow cover variability index mentioned across the globe. The duration of frozen ground is arguably one of the most biologically appropriate definitions of winter length, especially for vegetation. The duration of frozen ground with and without snow indices are again approximating subnivium conditions, which provide critical insulation against freezing temperatures for soil organisms, plants, and animals.
(2) Global Winter Habitat Indices derived from satellite data (25-km) by Likai Zhu: frozen season length (top-left), duration of frozen ground without (bottom-left) and with (bottom-right) snow, and snow cover variability (top-right). Calculated annually from 2000/01 – 2018/19.
To assess the potential of the WHIs for ecological research and predicting biodiversity patterns, David is working with collaborators at UW (Ben Zuckerberg, Jon Pauli, and Spencer Keyser) and the Cornell Lab of Ornithology (Daniel Fink) to model relationships between the WHIs and species richness for birds estimated from eBird data. The results have been promising: all of the WHIs have shown strong relationships with species richness patterns for birds across the US. Expectedly, measures of winter length (i.e., snow season length and duration of frozen ground) have a negative relationship with species richness across all taxa: areas with long winters have fewer species relative to those with shorter winters. The snow cover variability and frozen ground without snow indices, on the other hand, have more complex, nonlinear relationships with bird species richness. Generally, snow cover variability has had a positive relationship with species richness: areas with higher snow cover variability have more bird species than those with lower variability.
These results highlight the great potential and promise of the WHIs for ecological research, biodiversity assessments, and conservation planning. David and Likai continue to derive the WHIs from new datasets in hopes of maximizing the spatial and temporal resolutions available to researchers, with the latest being the 30-meter harmonized Landsat 8-Sentinel 2 dataset. Given the rapid rate of winter climate change, they hope others will utilize the WHIs in their winter ecology studies and species distribution models to inform conservation strategies moving forward.
Peatland fires can be some of the most catastrophic and expensive fires globally, including in the Great Lakes Region where wildfires in the mid-2000s burned through tens of thousands of acres in Michigan peatlands, and in 1976, when the Seney Fire burned for months underground consuming peat. Deep burning peatland fires can lead to the conversion and the permanent loss of peatlands, which are important refugias for plant and animal species. Such fires, however, have been infrequent since European settlement, and are assumed to have occurred only every few centuries or millennia. We investigated fire in peatlands to determine if historically fire in peatlands was indeed infrequent and severe.
Map of three peatland sites sampled across the Great Lakes Region in Wisconsin and Michigan.
Between 2017 and 2019, we collected samples from 220 fire-scarred trees in three peatlands within the Hiawatha and Chequamegon-Nicolet National Forests in the Great Lakes Region. Our sites were poor fen peatlands intermixed with dry to dry-mesic northern forests. We identified 141 fire years from 1548 to 1955. Prior to 1955 fires were historically frequent across our peatland sites occurring on average every 7 to 34 years depending on the site sampled. By using fire-scarred tree samples to investigate historical fire, we could identify fires that were likely low- to moderate-severity because fire scars are indicative of more frequent, low-severity fire whereas high-severity fires would have consumed most trees rather than just scarring them. We did not detect any fires after 1955, corresponding to the period when fire suppression efforts of the United States Forest Service became increasingly effective.
Fire-scarred tree sample including fires in 1734, 1755, and 1792 and the corresponding position (dormant or latewood) of the fire scar in individual tree rings.
Historically, fire was frequent in the peatlands we investigated and our research warrants rethinking of the role of fire in these ecosystems. Such a rethinking could have major implications on how managers use prescribed fire and mitigate wildfire risk in peatlands. Fire suppression since the mid-1900s in Great Lakes’ peatlands likely has unintended consequences including the encroachment of undesirable vegetation into open bogs and fens that were historically maintained by frequent fires over hundreds of years. Furthermore, it is possible that by removing frequent fire from peatlands in the Great Lakes Region these systems could become more prone to more severe, catastrophic fires that burn deep consuming peat soils. A possible mechanism for such a shift may be encroaching vegetation drawing more water from peat soils making them more susceptible to burning under severe drought conditions. Repeated prescribed burning under the more moderate conditions (e.g., surface vegetation of the peatlands is dry, but the peat soils are inundated with water) could be one approach to prevent encroachment and maintaining low-severity fire regimes. Fire management in peatlands of the Great Lakes Region needs to include rethinking the role of fire, considering it not just as a threat but also as a potential tool.
Collaborators: Jed Meunier (Wisconsin DNR) and Eric Rebitzke (US Forest Service)
Habitat connectivity is essential to facilitate species movement across fragmented landscapes, but hard to achieve at broad scales. The enforcement of existing land use policies could improve habitat connectivity, while providing legal support for implementation. Our goal was to evaluate how forest connectivity is affected if forests are restored according to existing riparian buffer regulations in Chile. We simulated forest restoration within 30 and 200 m of rivers in 99 large watersheds, following two sections of the forest regulation. We mapped habitat for two model forest species that have different minimum habitat sizes (15 and 30 ha), and for each we identified forest habitats and corridors using image morphology analysis. To quantify change in connectivity, we used a network graph index, the Relative Equivalent Connected Area. We found that both 30- and 200-m riparian buffers could have a positive effect on habitat connectivity. The 200-m buffers increased connectivity the most where forest cover was 20–40% (40% mean increase in connectivity index), while the 30-m buffers increased connectivity the most where forest cover was 40–60% (30% mean increase in connectivity index). The effect of riparian restoration scenarios was similar for both model species, suggesting that effective implementation of existing forest regulation could improve connectivity for fauna with a range of minimum habitat size requirements. Our findings also suggest that there is some flexibility in the buffer sizes that, if restored, would increase habitat connectivity. This flexibility could help ease the social and economic cost of implementing habitat restoration in productive lands.
Oceanic islands are important habitats for many endemic species. Global conservation assessments, however, are too coarse to characterize areas of high human influence or landscape connectivity at a resolution that is useful for conservation planning on most islands. Our goal was to identify landscape elements that are essential for the maintenance of structural connectivity among natural habitat patches on islands. Using the Caribbean island of Puerto Rico as a case study, our specific objectives were to: (1) develop a map of the human footprint, and (2) characterize the connectivity of patches exhibiting low human modification that structurally connect the island’s ecological network. We used the human footprint as a measure of impediments to connectivity among Puerto Rico’s natural areas using network analysis. We found that more than half of Puerto Rico’s current land surface had a low human footprint (56%), but that coastal areas were highly affected by human use (82%). Puerto Rico possesses a compact network of natural areas, with a few patches in the interior mountains critical to structural connectivity. The number of isolated patches is very high; more than 60% of the patches were 2000 m or more apart. Identifying sites that are key hubs to connectivity on islands and ensuring they remain undeveloped is one strategy to balance land use and conservation, and to facilitate the persistence of endemic species. We show here how to improve general conservation assessment methods to be more relevant for islands. There is potential to support an interconnected network of natural areas that promotes landscape connectivity in Puerto Rico among noncoastal habitats, because the human activities are concentrated along the coast whereas the interior mountain range has a relatively low human footprint.
Marine turtles may respond to projected climatic changes by shifting their nesting range to climatically suitable areas, which may
result in either increased exposure to threats or fewer threats. Therefore, there is the need to identify whether habitat predicted to
be climatically suitable for marine turtle nesting in the future will be affected by future threats and hinder marine turtles’ ability to
adapt. We modelled the geographic distribution of climatically suitable nesting habitat for marine turtles in the USA under future
climate scenarios, identified potential range shifts by 2050, determined impacts from sea-level rise, and explored changes in
exposure to coastal development as a result of range shifts. Overall nesting ranges of marine turtle species were not predicted to
change between the current and future time periods, except for the northern nesting boundaries for loggerhead turtles. However,
declines in climatically suitable nesting grounds were predicted; loggerhead turtles will experience the highest decreases (10%) in
climatically suitable habitat followed by green (7%) and leatherback (1%) turtles. However, sea-level rise is projected to inundate
78–81% of current habitat predicted to be climatically suitable in the future, depending on species and scenario. Nevertheless,
new beaches will also form, and suitable nesting habitat could be gained, with leatherback turtles potentially experiencing the
biggest percentage gain in suitable habitat.
