Page 52 - GIS for Science, Volume 3 Preview

CASE STUDY: FEDERAL LAND MANAGEMENT
The federal Bureau of Land Management (BLM), whose mission is “to sustain the health, diversity, and productivity of public lands for the use and enjoyment of present and future generations,” manages almost 250 million acres, more than 10% of the surface lands of the United States, the vast majority of which are in 13 western states. These lands are largely designated as multiple use, which charges the agency with balancing grazing, mineral and fossil fuel extraction, hunting, fishing, and off-road vehicle use with the conservation of species and their habitats. To do so efficiently and effectively, BLM must understand with high spatial precision where potential conflicts may exist among these diverse resource management goals.
New Mexico is a case in point. Recent years have seen an enormous increase in requests for oil and gas leases in areas of the state that are also known to harbor rare and imperiled plant species whose geographic distributions are poorly known. BLM staff review these requests, balancing the need for national energy production against the importance of preventing decline of native ecosystems and the species that inhabit them.
metadata, so the process is transparent and reproducible. The model results can be used to guide field biologists to efficiently survey the areas of highest probability of suitable habitat. The results of field surveys in turn provide new presence and absence data for the next round of models, resulting in increased precision in model outputs. Approached this way, habitat modeling is a continuous, dynamic process, and the distribution information that guides decisions about our nation’s most imperiled species represents the best available science.
The imperiled Justicia wrightii, or Wright’s water-willow, is known to occur only in a few locations in the shortgrass grasslands and shrublands of southeastern New Mexico and west-central Texas. Oil and gas development is extensive where it occurs on BLM lands in New Mexico.
Prior to habitat modeling efforts, information on the distribution of Justicia wrightii in New Mexico was limited
to broad-range maps or field observations. Without adequate information on the species distribution, assessing conserva- tion status or mitigating threats was difficult.
To address this challenge, the New Mexico office of the BLM worked with NatureServe and the New Mexico Natural Heritage Program to develop habitat models for six plant species that could be impacted by oil and gas development, including a rare water- willow (Justicia wrightii), a federally endangered milkvetch (Astragalus gypsodes), and a critically imperiled species of flax (Linum allredii) that was only described as a new species in 2011. Existing distribution information for the plants was fragmentary, but it was known that their habitat might overlap with specific areas targeted for new oil and gas development.
Justicia wrightii
Field surveys that confirm presence or absence of the species are the most verifiable tool for decision-making—but they can also be time-consuming, needle-in-a- haystack efforts across vast desert expanses. Using documented observations for these species, NatureServe developed habitat models and used statistically defined thresholds to identify areas with low, medium, and high probability of suitable habitat. Using these 30-m resolution maps as a guide, botanists from the New Mexico Natural Heritage Program designed a structured field survey protocol and went on an imperiled plant species treasure hunt.
For several of these species, the hunt was successful in finding new, previously undocumented populations. Presence and absence data recorded during these field surveys were used together with refined environmental predictor inputs, including detailed soils maps, to develop a second-generation set of models for these rare plants.
This example highlights perhaps the most important and useful feature of habitat modeling as a tool for supporting species management and conservation. Beginning with available information on confirmed occurrence of a target species and high- resolution digital data on variables such as soils, climate, and topography, the modeling process produces a spatially explicit, testable hypothesis of the probability of suitable habitat. Choices regarding model parameters can be recorded in the
40
GIS for Science