Areas of high connectivity value and where they overlap with areas of high residential development and high risk of solar development, developed for the Washington Habitat Connectivity Action Plan.

This layer combines data from three analyses:

1) Areas where high connectivity values were aggregated into homogenous patches.

We used the landscape connectivity surface to identify areas with high connectivity values that were aggregated into homogenous patches. To achieve this we applied a cost-kernel approach to a set of source points which were selected probabilistically proportional to the landscape connectivity surface values. The kernel analysis penalized spread as a nonlinear function of connectivity value, such that the kernel spreads farther through areas of continuously high connectivity value and spreads shorter distances in areas of low connectivity value. This produced a kernel density surface showing areas of high connectivity value that are aggregated across the landscape.

2) Areas of both high connectivity value and residential development risk.

To identify areas with concentrated extents of high connectivity value which are also facing significant threat from potential housing and residential development, we applied kernel density analyses following the methodology detailed in the preceding section by adjusting the cost surface to account for the development.

To specifically quantify areas under high development pressure, we employed the Mann-Kendall Index. This non-parametric test was computed to detect statistically significant monotonic trends in a 30-year Landsat satellite image time series. For each landscape unit, the Mann-Kendall Index yielded a value where positive indices indicated an increasing trend in development pressure. We rescaled the positive index values to a range of 0 to 1, with 1 representing the highest development pressure.

This rescaled development pressure layer was subsequently integrated with areas of relatively high connectivity. To achieve this, we selected portions of the original connectivity surface with values exceeding the 30th percentile (connectivity value > 8) and rescaling it to a 0-to-1 range, where 1 denoted the highest connectivity within this subset.

Finally, we multiplied the rescaled development pressure layer by the rescaled high-connectivity layer. The resulting surface, ranging from 0 to 1, represented the spatial congruence of high development threat and high connectivity, signifying areas of heightened vulnerability.

We used the multiplication layer as the base to generate source locations and ’cost surface’ to fit and compute kernel density. We set the kernel cost-weighted distance at 79,200 – an equivalent of 15 mi in uniform landscape with the lowest cost defined by highest development and connectivity values

3) Areas of both high connectivity value and solar development risk.

To identify focal connectivity areas with a high threat of solar development we used The Solar Development Suitability Model for Columbia Plateau created by a mapping group for the Least-Conflict Solar Project managed by Washington State University Energy Program (https://www.energy.wsu.edu/RenewableEnergy/LeastConflictSolarSiting.aspx). The model was built using Environmental Evaluation Modeling System (EEMS) developed by the Conservation Biology Institute (CBI). EEMS is a flexible, data-driven framework that uses fuzzy logic to integrate various spatial data layers and assess complex environmental and planning questions. It allows for the combination of different types of data (e.g., ecological, infrastructure, land use) to produce a suitability map. The logic of The Solar Development Suitability Model aimed at depicting relative physical suitability for utility scale passive solar development. At a high level of this hierarchical model, high development suitability was defined by characteristics such as terrain (slope and aspect) and soil conditions, high proximity to existing road and transmission infrastructure, and to a lesser degree potential hazards (i.e., wildfire and earthquakes).

The values of The Solar Development Suitability Model ranged from -1 (highly unsuitable) to 1 (highly suitable). We extracted only the positive values by setting all negative values to 0. We then multiplied this layer by the rescaled (0-1) connectivity value surface, representing areas with relatively high connectivity (portions of the original connectivity surface with values exceeding the 30th percentile). The resulting layer assigned higher values to areas of high connectivity threatened by solar development.

We then used this final multiplication layer to compute a kernel density surface. Similarly to the residential development analyses, we applied a kernel cost-weighted distance of 79,200.

Areas of high connectivity value and where they overlap with areas of high residential development and high risk of solar development, developed for the Washington Habitat Connectivity Action Plan.

This layer combines data from three analyses:

1) Areas where high connectivity values were aggregated into homogenous patches.

We used the landscape connectivity surface to identify areas with high connectivity values that were aggregated into homogenous patches. To achieve this we applied a cost-kernel approach to a set of source points which were selected probabilistically proportional to the landscape connectivity surface values. The kernel analysis penalized spread as a nonlinear function of connectivity value, such that the kernel spreads farther through areas of continuously high connectivity value and spreads shorter distances in areas of low connectivity value. This produced a kernel density surface showing areas of high connectivity value that are aggregated across the landscape.

2) Areas of both high connectivity value and residential development risk.

To identify areas with concentrated extents of high connectivity value which are also facing significant threat from potential housing and residential development, we applied kernel density analyses following the methodology detailed in the preceding section by adjusting the cost surface to account for the development.

To specifically quantify areas under high development pressure, we employed the Mann-Kendall Index. This non-parametric test was computed to detect statistically significant monotonic trends in a 30-year Landsat satellite image time series. For each landscape unit, the Mann-Kendall Index yielded a value where positive indices indicated an increasing trend in development pressure. We rescaled the positive index values to a range of 0 to 1, with 1 representing the highest development pressure.

This rescaled development pressure layer was subsequently integrated with areas of relatively high connectivity. To achieve this, we selected portions of the original connectivity surface with values exceeding the 30th percentile (connectivity value > 8) and rescaling it to a 0-to-1 range, where 1 denoted the highest connectivity within this subset.

Finally, we multiplied the rescaled development pressure layer by the rescaled high-connectivity layer. The resulting surface, ranging from 0 to 1, represented the spatial congruence of high development threat and high connectivity, signifying areas of heightened vulnerability.

We used the multiplication layer as the base to generate source locations and ’cost surface’ to fit and compute kernel density. We set the kernel cost-weighted distance at 79,200 – an equivalent of 15 mi in uniform landscape with the lowest cost defined by highest development and connectivity values

3) Areas of both high connectivity value and solar development risk.

To identify focal connectivity areas with a high threat of solar development we used The Solar Development Suitability Model for Columbia Plateau created by a mapping group for the Least-Conflict Solar Project managed by Washington State University Energy Program (https://www.energy.wsu.edu/RenewableEnergy/LeastConflictSolarSiting.aspx). The model was built using Environmental Evaluation Modeling System (EEMS) developed by the Conservation Biology Institute (CBI). EEMS is a flexible, data-driven framework that uses fuzzy logic to integrate various spatial data layers and assess complex environmental and planning questions. It allows for the combination of different types of data (e.g., ecological, infrastructure, land use) to produce a suitability map. The logic of The Solar Development Suitability Model aimed at depicting relative physical suitability for utility scale passive solar development. At a high level of this hierarchical model, high development suitability was defined by characteristics such as terrain (slope and aspect) and soil conditions, high proximity to existing road and transmission infrastructure, and to a lesser degree potential hazards (i.e., wildfire and earthquakes).

The values of The Solar Development Suitability Model ranged from -1 (highly unsuitable) to 1 (highly suitable). We extracted only the positive values by setting all negative values to 0. We then multiplied this layer by the rescaled (0-1) connectivity value surface, representing areas with relatively high connectivity (portions of the original connectivity surface with values exceeding the 30th percentile). The resulting layer assigned higher values to areas of high connectivity threatened by solar development.

We then used this final multiplication layer to compute a kernel density surface. Similarly to the residential development analyses, we applied a kernel cost-weighted distance of 79,200.