 My name is Brad Eichelberger. I'm a GIS analyst and landscape ecologist with the Natural Capital Project, and I'll be talking to you about the Invest Habitat Quality Model. So biodiversity is intimately linked to the production of environmental services but is not considered an ecosystem service per se. So we treat it as an independent attribute of natural systems with its own intrinsic value. Therefore, it can be used to assess overlaps and trade-offs with other ecosystem services. To understand the patterns of distribution and richness across the landscape, individually and in aggregate, it is necessary to map the range of consequences of elements, for example, species, communities, and habitats. It is important to understand threats and the management affects the persistence of these elements in order to design appropriate conservation strategies and encourage resource management that maximizes biodiversity in those areas. So the main question we'll address is how do threats affect habitat and whether areas are better able to support and maintain biodiversity. It is important to note that both the Habitat Quality Model and the Habitat Risk Assessment Model, also known as HRA, can be used to examine how risk posed by human activities varies across the landscape or ski scape and identify the locations where risk is highest or lowest. The two models are mathematically similar but differ in a few key ways. Habitat Quality was developed with terrestrial systems in mind, whereas HRA is applicable at both marine and terrestrial systems. Habitat Quality is a bit more simplified in the sense that it does not directly include attributes such as resilience, recruitment, and other factors. The risk assessment framework allows HRA model results to be interpreted along exposure and consequences axisies in a sense that helps users explore which types of management strategies are likely to mis-effectively reduce risk. The HRA model incorporates, again, those additional criteria such as resilience, recruitment, and other factors. This schematic depicts the general overview of the model. We have a species that we are interested in modeling, in this case Northern Spotted Isle. We have areas designated as habitat, forced in this case, and non-habitat such as this logged area. Finally, we have threats such as roads and urbanization that degrade the habitat and the impact of these threats decay over distance. As with any model, there are few limitations and caveats we need to consider before running the Invest Habitat Quality model. The structure of the model assumes threats are additive, whereas in reality some interactions may be multiplicative. The habitat boundaries are user-defined and care should be taken to ensure that the habitat boundaries incorporate all the areas of interest for the species. And lastly, the model does not account for connectivity, patch size, or habitat resilience directly. However, these factors should be accounted for by adjusting the habitat quality and sensitivity scores which we'll talk about in one second. There are two main sets of input data for the model. Threats to habitat are species. In this example, we have roads and urbanization, the relative weight of the impact for each threat on the habitat, and these weights can range from one at the highest weight to zero at the lowest weight. We also look at the spatial impact of the threat in terms of a maximum distance which will decline to zero at this maximum distance. By default, the distance decay function of the threat is assumed to be exponential, which is indicative of a pattern seen in ecology. The second set of data corresponds to the land use land cover. Pixels are assigned a suitability score which ranges from zero, which indicates non-habitat, to one, which indicates ideal habitat. We also consider the relative sensitivity of each habitat type to each threat. These values range from zero to one, where one represents high sensitivity to a threat and zero represents no sensitivity. And lastly, we also consider areas with accessibility. So things such as strict nature reserves, well-protected private lands are assigned some number less than one, where polygons with maximum accessibility, such as extractive reserves, are assigned a value of one. There are three main outputs for the model. First, we have habitat quality, which represents habitat quality on the current landscape. Higher numbers indicate better habitat quality. These are the distribution of habitat quality across the rest of the landscape. Areas on the landscape that are not habitat get a quality score of zero. The quality score is unitless and does not refer to any particular biodiversity measure. Our second model output is habitat degradation. This represents a relative level of habitat degradation on the current landscape. A high score in this grid means habitat degradation in the cell is high relative to all the other cells. And lastly, we have habitat relative abundance. And this represents the relative number of the habitat cells that occur within the landscape as relative to the baseline raster. In summary, this model can be used to help understand the spatial effects of threats on habitats and which areas are most affected. Resource manager can then use this model to better understand how management affects species habitat and the tradeoffs with other ecosystem services.