Spatial Analysis For Conservation and Sustainability
Land Use
Land use change is currently the largest threat to biodiversity, and exacerbates the detrimental effects of climate change. We are interested in novel types of land use change, such as housing growth in the WUI, and widespread land abandonment after socioeconomic shocks, and how such changes affect biodiversity.
Land cover change is one of the major contributors to global change, but long-term, broad-scale, detailed and spatially explicit assessments of land cover change are largely missing, although the availability of historical maps in digital formats is increasing. The problem often lies in efficiency of analyses of historical maps for large areas. Our goal was to assess different methods to reconstruct land cover and land use from historical maps to identify a time-efficient and reliable method for broad-scale land cover change analysis. We compared two independent forest cover reconstruction methods: first, regular point sampling, and second, wall-to-wall mapping, and tested both methods for the Polish Carpathians (20,000 km2 ) for the 1860s, 1930s and 1970s. We compared the two methods in terms of their reliability for forest change analysis, relative to sampling error, point location and landscape context including local forest cover, area of the spatial reference unit and forest edge-to-core ratio. Our results showed that the point-based analysis overestimated forest cover in comparison to wall-to-wall mapping by 1e3%, depending on the mapping period. The reasons for the differences were mainly the backdating approach and map generalisation rather than the point grid position or sampling error. When we compared forest cover trajectories over time, we found that the point-based reconstruction captured forest cover dynamics with a comparable accuracy to the wall-to-wall mapping. More broadly, our assessment showed that historical maps can provide valuable data on long-term land cover trends, and that point-based sampling can be an efficient and accurate way to assess forest area and change trends. We suggest that our point-based approach could allow land cover mapping across much of Europe starting in the 1800s. Our findings are important because they suggest that land cover change, a key component of global change, can be assessed over large areas much further back in time than it is commonly done. This would allow to truly understand path dependencies, land use legacies, and historical drivers of land cover change
Historic land use can exert strong land-use legacies, i.e., long-lasting effects on ecosystems, but the importance of land-use legacies, alongside other factors, for subsequent forest-cover change is unclear. If past land use affects rates of forest disturbance and afforestation then this may constrain land use planning and land management options, and legacies of current land management may constrain future land use. Our goal was to assess if and how much land-use legacies affect contemporary forest disturbance, and the abundance of different forest types in the Carpathian region in Eastern Europe (265,000 km2 , encompassing parts of Poland, Slovakia, Ukraine, Romania, Hungary, and Czech Republic). We modeled contemporary forest disturbance (based on satellite image analysis from 1985 to 2010) as a function of historic land use (based on digitized topographic maps from 1860 and 1960). Contemporary forest disturbance was strongly related to historic land use even when controlling for environmental, accessibility and socio-political variation. Across the Carpathian region, the odds of forest disturbance were about 50% higher in areas that were not forested in 1860 (new forests) compared to areas that were forested then (old forests). The forest disturbance in new forests was particularly high in Poland (88% higher odds), Slovakia (69%) and Romania (67%) and persisted across the entire range of environmental, accessibility and socio-political variation. Reasons for the observed legacy effects may include extensive plantations outside forest ranges, predominantly spruce, poplar, and black locust, which are prone to natural disturbances. Furthermore, as plantations reach harvestable age of about 70 years for pulp and 120 year for saw-timber production, these are likely to be clear-cut, producing the observed legacy effects. Across the Carpathians, forest types shifted towards less coniferous cover in 2010 compared to the 1860s and 1960s likely due to extensive historic conifer harvest, and to recent natural disturbance events and clear-cuts of forest plantations. Our results underscore the importance of land-use legacies, and show that past land uses can greatly affect subsequent forest disturbance for centuries. Given rapid land use changes worldwide, it is important to understand how past legacies affect current management and what the impact of current land management decisions may be for future land use.
The wildland–urban interface (WUI) is the area where houses meet or intermingle with undeveloped wildland vegetation. The WUI is thus a focal area for human– environment conflicts, such as the destruction of homes by wildfires, habitat fragmentation, introduction of exotic species, and biodiversity decline. Our goal was to conduct a spatially detailed assessment of the WUI across the United States to provide a framework for scientific inquiries into housing growth effects on the environment and to inform both national policymakers and local land managers about the WUI and associated issues. The WUI in the conterminous United States covers 719 156 km2 (9% of land area) and contains 44.8 million housing units (39% of all houses). WUI areas are particularly widespread in the eastern United States, reaching a maximum of 72% of land area in Connecticut. California has the highest number of WUI housing units (5.1 million). The extent of the WUI highlights the need for ecological principles in land-use planning as well as sprawl-limiting policies to adequately address both wildfire threats and conservation problems.
Agricultural land abandonment is a common land-use change, making the accurate mapping of both location and timing when agricultural land abandonment occurred important to understand its environmental and social outcomes. However, it is challenging to distinguish agricultural abandonment from transitional classes such as fallow land at high spatial resolutions due to the complexity of change process. To date, no robust approach exists to detect when agricultural land abandonment occurred based on 30-m Landsat images. Our goal here was to develop a new approach to detect the extent and the exact timing of agricultural land abandonment using spatial and temporal segments derived from Landsat time series. We tested our approach for one Landsat footprint in the Caucasus, covering parts of Russia and Georgia, where agricultural land abandonment is widespread. First, we generated agricultural land image objects from multi-date Landsat imagery using a multiresolution segmentation approach. Second, we estimated the probability for each object that agricultural land was used each year based on Landsat temporal-spectral metrics and a random forest model. Third, we applied temporal segmentation of the resulting agricultural land probability time series to identify change classes and detect when abandonment occurred. We found that our approach was able to accurately separate agricultural abandonment from active agricultural lands, fallow land, and re-cultivation. Our spatial and temporal segmentation approach captured the changes at the object level well (overall mapping accuracy = 97 ± 1%), and performed substantially better than pixel-level change detection (overall accuracy = 82 ± 3%). We found strong spatial and temporal variations in agricultural land abandonment rates in our study area, likely a consequence of regional wars after the collapse of the Soviet Union. In summary, the combination of spatial and temporal segmentation approaches of time-series is a robust method to track agricultural land abandonment and may be relevant for other land-use changes as well.
Land-use change significantly contributes to biodiversity loss, invasive species spread, changes in biogeochemical cycles, and the loss of ecosystem services. Planning for a sustainable future requires a thorough understanding of expected land use at the fine spatial scales relevant for modeling many ecological processes and at dimensions appropriate for regional or national-level policy making. Our goal was to construct and parameterize an econometric model of land-use change to project future land use to the year 2051 at a fine spatial scale across the conterminous United States under several alternative land-use policy scenarios. We parameterized the econometric model of land-use change with the National Resource Inventory (NRI) 1992 and 1997 land-use data for 844 000 sample points. Land-use transitions were estimated for five land-use classes (cropland, pasture, range, forest, and urban). We predicted land-use change under four scenarios: business-as-usual, afforestation, removal of agricultural subsidies, and increased urban rents. Our results for the business-as usual scenario showed widespread changes in land use, affecting 36% of the land area of the conterminous United States, with large increases in urban land (79%) and forest (7%), and declines in cropland (16%) and pasture (13%). Areas with particularly high rates of landuse change included the larger Chicago area, parts of the Pacific Northwest, and the Central Valley of California. However, while land-use change was substantial, differences in results among the four scenarios were relatively minor. The only scenario that was markedly different was the afforestation scenario, which resulted in an increase of forest area that was twice as high as the business-as-usual scenario. Land-use policies can affect trends, but only so much. The basic economic and demographic factors shaping land-use changes in the United States are powerful, and even fairly dramatic policy changes, showed only moderate deviations from the business-as-usual scenario. Given the magnitude of predicted land-use change, any attempts to identify a sustainable future or to predict the effects of climate change will have to take likely land-use changes into account. Econometric models that can simulate land-use change for broad areas with fine resolution are necessary to predict trends in ecosystem service provision and biodiversity persistence.
Land-use change significantly contributes to biodiversity loss, invasive species spread, changes in biogeochemical cycles, and the loss of ecosystem services. Planning for a sustainable future requires a thorough understanding of expected land use at the fine spatial scales relevant for modeling many ecological processes and at dimensions appropriate for regional or national-level policy making. Our goal was to construct and parameterize an econometric model of land-use change to project future land use to the year 2051 at a fine spatial scale across the conterminous United States under several alternative land-use policy scenarios. We parameterized the econometric model of land-use change with the National Resource Inventory (NRI) 1992 and 1997 land-use data for 844 000 sample points. Land-use transitions were estimated for five land-use classes (cropland, pasture, range, forest, and urban). We predicted land-use change under four scenarios: business-as-usual, afforestation, removal of agricultural subsidies, and increased urban rents. Our results for the business-as usual scenario showed widespread changes in land use, affecting 36% of the land area of the conterminous United States, with large increases in urban land (79%) and forest (7%), and declines in cropland (16%) and pasture (13%). Areas with particularly high rates of landuse change included the larger Chicago area, parts of the Pacific Northwest, and the Central Valley of California. However, while land-use change was substantial, differences in results among the four scenarios were relatively minor. The only scenario that was markedly different was the afforestation scenario, which resulted in an increase of forest area that was twice as high as the business-as-usual scenario. Land-use policies can affect trends, but only so much. The basic economic and demographic factors shaping land-use changes in the United States are powerful, and even fairly dramatic policy changes, showed only moderate deviations from the business-as-usual scenario. Given the magnitude of predicted land-use change, any attempts to identify a sustainable future or to predict the effects of climate change will have to take likely land-use changes into account. Econometric models that can simulate land-use change for broad areas with fine resolution are necessary to predict trends in ecosystem service provision and biodiversity persistence.
People enjoy building houses in beautiful places where they are surrounded by the beauty of nature. Unfortunately, when wildfires rage through forests, these homes are often caught in the fire’s path. As more and more people attempt to enjoy the amenities of building a home in sparsely populated areas, communities increasingly face tough decisions whether to pay for protecting these homes from wildfires that destroy property and take lives. Wildfire costs are not trivial. During the twelve-years from 1999-2011, an average of 1,354 houses were destroyed and approximately $2 billion was spent fighting wildfires, annually. Ideally, communities would have information to help predict how wildfires spread and how to minimize the number of houses lost during wildfires. Unfortunately, a lot of basic information about what happens to a community after a wildfire rips though it is unknown. Patricia Alexandre and her colleagues recently published a study that makes a first step towards describing what happens to communities across the country following wildfire events. While their results suggest that the conventional wisdom that rebuilding always happens has little support and how much is rebuilt varies across the country, they were surprised to find that new housing constructed in burned areas was happening at higher rates than rebuilding, and often at higher rates than in surrounding non-fire areas, adding complexity to the discussion.
Example of a rebuilt building situation when using Google Earth
To be clear, Alexandre’s research is not intended to answer whether people should rebuild following a wildfire, but to provide a snapshot of what the patterns were within affected communities across the country following recent fires. This is an important step to take to see whether patterns are consistent across a large scale and provides a dataset to begin drawing conclusions from observed rebuilding patterns. To do this work, Alexandre refined a method utilizing historical images available on Google Earth and then recruited help from students to go through and hand-digitize structures present before a fire as well as all structures that had burned, and consequently been rebuilt within five years following a fire. Nationally, the team found rebuilding rates averaged 25%, with much higher rates in the western states. For example, rates in California approached 70% of structures rebuilt following a fire. The surprising results from Alexandre’s work is first that not all burned communities are re-building within five years following a fire, and second that new buildings were constructed in burned areas at similar or even higher rates. These results indicate that communities are not just replacing homes lost to wildfires, but many are putting new homes into burned areas.
Percentage of burned buildings per fire (fires occurred between 2000 and 2005)
Alexandre offers multiple reasons that homeowners may build, and rebuild, in burned areas, which are inherently fire prone. One reason is that homeowners may find the value they get from living in fire-prone areas worth the fire risk, some insurance policies require rebuilding in the same spot following a fire, and many homeowners do not have the finances to relocate to an area with lower fire risk. While the reasons to build and rebuild in fire prone areas likely vary widely across the country, Alexandre’s research provides a valuable baseline to evaluate future policies or practices that communities might use to mitigate wildfire damages. Whether it’s mandating that new or rebuilt structures be constructed with safer materials, or prohibiting rebuilding in burned areas, the best way to evaluate the efficacy of these policies is to compare them to the housing patterns before and after fire events. Alexandre’s research allows that comparison to take place and hopefully inform local initiatives that could save property, money and lives.”
Figure 1. Location of the Sierra Gorda Biosphere reserve in central Mexico
In central Mexico, there is a geographically and biologically unique space featuring a mosaic of a varied diversity of flora and fauna, the Sierra Gorda Biosphere Reserve. Yet similar to other reserves in the world, human and climate drivers are changing the unique conditions of this area at an accelerated rate (Figure 2). However, this reserve remains home to important protected and endangered species that are struggling to find optimal habitat conditions elsewhere, thus it is important to identify potential areas that are suitable for keep these animal populations.Jaguar (Panthera onca) is one of these endangered species, and essential in the faunal community because of its position as a top predator (Figure 2).
Figure 2. Jaguar captured in a camera trap by Grupo Ecologico Sierra Gorda
Furthermore, the presence of jaguars is an indicator of the health of the ecosystems and by conserving its habitat we can also protect other species. However, is very difficult to allocate efforts to large areas and therefore we need to identify and prioritize the most important zones capable to host jaguars.In the SILVIS lab, Carlos Ramirez Reyes, created a potential species distribution model for jaguars based on presence data and using factors than can affect their distribution such as topography, landcover and precipitation among others (Figure 3).
Figure 3. Potential distribution for jaguars in the Sierra Gorda Biosphere Reserve
Based on this model, he identified areas inside within and outside the reserve that could host jaguars. Additionally, by adding connectivity parameters to the models (Figure 4), he evaluated if habitat connectivity actually improves the potential distribution model for the reserve. The goal was to find patches of potential habitat that are critical for connectivity of the entire reserve.
Figure 4. Relative importance of potential habitat patches to maintain the connectivity of the system
With his research Carlos hopes to gain better knowledge for the habitat requirements of the jaguar in the region. Ultimately, the project should serve to inform nongovernmental organizations and government agencies with interests in the reserve to make decisions on how to distribute resources for the management of the species and prioritize areas that should be protected. Furthermore, this project includes a new method for obtaining potential areas through the use of landscape connectivity. Similar projects that aim to model potential species distribution can benefit from this technique given that landscape connectivity is also important for animal dispersal and gene flow in fragmented landscapes. “
Max made a technological contribution to both the fields of wildlife ecology, and parks & recreation by developing a device to measure how heavily trails are used. His goal was to quantify both group size and frequency of groups (groups/hour) along a given trail, but the available solutions were more than his research budget could manage. Having someone count hikers all day along several trails required more personnel than was practical. Meanwhile, he worried that sampling use in small time periods would provide representative data, because trail use varies throughout the day. The idea to use an automatic sensor was desirable, but the options on the market were too expensive. So he collaborated with someone with technical expertise to invent a tool that met his needs.
Components of the Trail Monitor inside a protective weather-proof box
The solution was found in open source software and DIY hardware. First, he acquired a passive infrared (PIR) sensor that can detect warm-bodied objects that passed by (these are the same types of sensors that control automatic light switches by detecting when someone walks into a room). Then, he connected this sensor to an Arduino Uno board (http://www.arduino.cc/) that supports open source software. The board receives input from the sensor, and can be controlled by a user-written script. This is connected to a data logging shield (http://www.adafruit.com/product/1141) which contains a clock and an SD card to store data. Then, the data can be imported Excel sheet. Max used pivot tables to translate the sensor’s detections into his variables of interest. For example, the duration of time the sensor is activated can be used as an index of the number of people in a group passing by.
Installing a trail monitor along a trail
Max’s invention is a great alternative to what’s commercially available, in part due to the price point: one of Max’s units costs less than $250, in contrast to commercially available counters that cost about $1000/unit. Also, Max’s device can be left out in the woods for about a week between battery replacement. Its relatively small size means it can be easily hidden, which makes it relatively safe from tampering. Thus, Max continues to produce technology that will likely be used by many researchers in the future! “
In an unusual twist, Paul Schilke’s interest in terrestrial birds has led him to study aquatic systems.
Map of study area (Chequamegon-Nicolet National Forest) with study sites and bodies of water.
Many aerial insectivore bird species, such as swallows and flycatchers, have been declining since the 1980s, but researchers aren’t sure why because little is known about how these birds use the resources around them. This guild is defined by its habit of capturing flying insects in midair, as opposed to the gleaner guild that picks insects off of substrates like leaves or twigs. Many of these flying insects begin their lifecycle in aquatic systems, so Paul thought that the differential decline in the aerial feeding guild might lie in the lakes and streams.
Using records from 317 locations within the Chequamegon-Nicolet National Forest in Northern Wisconsin, Paul compared presence of the aerial and gleaner insectivore guild members to estimated insect productivity in nearby lakes and streams, controlling for habitat differences (Figure 1). He estimated insect probability using a model from Bartrons et al. (2013), which used an extensive meta-analysis to determine the relationships between aquatic insect productivity and basic properties of lakes and streams such as temperature, surface area, and clarity. As expected given their feeding behavior, gleaners preferred forested habitats while aerial insectivores preferred more open areas. Interestingly, despite both guilds being insectivorous, aerial feeders demonstrated a strong preference for sites with higher insect inputs, while gleaners had no response (Figure 2).
Figure 2. Scatter plot of response of aerial insectivores (left) compared to response of gleaners (right) to estimated emergent aquatic insect inputs.
Paul hopes that a better understanding of the food resources of aerial insectivores can lead to better conservation measures, and hopefully reverse their long term decline. He will continue his work as a PhD student in the SILVIS lab.
Konrad Turlej, who brought his great expertise of remote sensing to SILVIS, enthusiastically started his PhD project in 2015 focusing on the mapping of tree species in Poland and in Wisconsin. The idea of the project is to map Polish and Wisconsin forests with 20-30 m resolution imagery. Konrad’s goal is ambitious: he wants to map not only where forests are, but also tree species with this medium resolution satellite data.
The reason why this is complicated is that a single tree is less than 30 mand there can be 2-4 trees species in a single pixel. Furthermore, in a satellite image, many tree species look very similar during the peak of summer. However, phenology varies greatly among tree species. Some trees have their leaves earlier, some ater. In the fall, some trees are losing leaves in September while others keep them till the first frost.
Forest Study Areas in Poland and Wisconsin
By analyzing satellite images for the entire growing season, one can analyzes so-called phenology curves. Over the course of a year, this curve looks different for different species because of phenological differences. This idea sounds quite promising but there is another challenge: Landsat, the source of 30 m imagery, provides only 1-2 images per months. In Poland, where leaves can fully come out in two weeks, this is not frequent enough to build a good phenological curve. This is why Konrad will combine Landsat data with imagery from other satellite sensors, including MODIS and Sentinel 2a.
Ultimately, the maps that Konrad is creating will be beneficial for several purposes. Describing and counting trees in the field takes a lot of time and money. Mapping maps of tree species from satellite images instead will save money and provide more timely information. With such maps, foresters can then estimate current situation on various species, amount of timber and its economic value, and eventually provide better management.
