Remote sensing is being used to promote the sustainability of agricultural and recreational opportunities in the Medicine Bow National Forest of Wyoming, USA.
Bromus tectorum also known as cheatgrass, is native to Asia, Africa and Europe, and was introduced to the United States in 1861. The increasing presence of cheatgrass in the U.S. is known to alter fire regimes and destroy native plant and animal habitat. Image Credit: Deeneg
On a quiet patch of land in the Medicine Bow National Forest (MBNF) in Wyoming, livestock consumed native grasses that dominated the landscape, and hikers trekked through those grasses to access tall aspen stands and striking views. There were mule deer sheltered in the aspen stands, and hunters who fed their families with those deer. But this ecosystem changed on June 27, 2012, when a lightning strike ignited the 98,000-acre Arapaho fire—the largest Wyoming wildland fire of that season.
Isn’t Fire a Natural Part of Most Ecosystems?
While periodic fires are critical to the health of the landscape, fires also create an ideal setting for the encroachment of invasive plant species such as cheatgrass (Bromus tectorum). Katie Haynes, a botanist for the MBNF, explained that cheatgrass is not only unsuitable forage for livestock and wildlife, but can change the fire regime to encourage a total takeover by the species.
ÛÏCheatgrass loves to burn and when it burns it spreads to even bigger areas, which is why we see it after firesÛ_ and it likes to burn almost every year,Û said Haynes, glancing at the carpet of cheatgrass flourishing in front of a charred aspen in the Arapaho burn area. Haynes described a landscape that could burn every 25-100 years historically. But with cheatgrass in the picture, the time between fires is drastically shorter.
This change in fire regimes decreases habitat for native species, diminishes natural ranchlands and threatens recreational opportunities within the burn area.
Farmers and land managers in and around MBNF have grown accustomed to changes on the landscape over time. Thus far, information passed from generation to generation combined with time-consuming field data collection has allowed recreation and ranching in MBNF to thrive. These ecosystems are often resilient in the face of change; however drought, climate change, and the introduction of non-native species may threaten the sustainability of recreational and agricultural opportunities in MBNF.
In response to these threats, the MBNF and the NASA DEVELOP program are looking to the sky and using remotely sensed data from NASA satellites to create economical solutions for non-native species treatment in the Arapaho fire burn area.
An artist’s rendering of the Landsat Data Continuity Mission (LDCM) in orbit. Landsat 8 images the entire Earth with multiple sensors every 16 days and provides free and open access to Earth observations for the benefit of society. Data from Landsat 8 has helped scientists produce some of the most detailed maps of the Earth’s surface to date. Image Credit: NASA
How are Satellites Used to Target Invasive Species Treatment?
Widespread aerial application of the herbicide imazapic is the MBNF’s current plan for cheatgrass treatment in the Arapaho burn area. The DEVELOP Wyoming Ecological Forecasting team joined forces with the MBNF, the U.S. Geological Survey (USGS), and Dr. Amanda West from the Natural Resources Ecology Lab at Colorado State University to map the spatial extent of cheatgrass cover in the site of the Arapaho Fire using satellite imagery as a cost-effective alternative to field data collection.
A cheatgrass presence probability map created by the DEVELOP Wyoming Ecological Forecasting team. Areas with a high probability of cheatgrass presence are represented in red/orange. These regions will be areas of focus in future cheatgrass mitigation efforts by the Medicine Bow National Forest. Image Credit: DEVELOP Wyoming Ecological Forecasting team 2015
Vegetation and topographic indices derived from NASA satellite imagery were combined with cheatgrass presence/absence field data collected by West and her team. These data were analyzed using the USGS Software for Assisted Habitat Modeling (SAHM). SAHM fit these data to models, which quantify the probability of cheatgrass occurring in each pixel. For example, if a species is known to grow in moist soil with no canopy on southern exposures below 8,000 feet, SAHM will calculate the probability of that species occurring in each pixel, based on how well that pixel’s characteristics match the stated criteria. The researchers used the outputs from this software to create a map predicting the probability of cheatgrass presence for the entire study area.
Haynes explained that knowing when to spray is as important as knowing where to spray. The most effective time to apply imazapic is during germination, or when seeds are just sprouting. The MODIS Global Subsetting Tool was used to estimate when cheatgrass is germinating in the study area. DEVELOP used the tool to produce a six-year time series of the Normalized Difference Vegetation Index and the Enhanced Vegetation Index for pixels with high probability of cheatgrass presence. Sharp increases in these indices suggest rapid vegetation growth. By knowing when these increases first occur, land managers may be able to more efficiently time aerial spraying, saving the MBNF time and money.
This image, derived from the MODIS Global Subsetting Tool, highlights changes in pre- and post-fire Enhanced Vegetation Index (EVI) values from an area of the Arapaho fire where cheatgrass was predicted to be present. Image Credit: DEVELOP Wyoming Ecological Forecasting Team 2015
Why Remote Sensing?
The threats of cheatgrass are severe, but there is little funding available for invasive species treatment projects. According to the 2008-2010 Invasive Species Action Plan for the Medicine Bow National Forest, the last inventory of invasive weed species in the area was completed in 1994. The report states that an updated inventory is not planned for the future due to lack of funding and resources. Methods such as those described in the Wyoming Ecological Forecasting team’s research combined with ground-truthing make quantifying areas of invasive species populations a feasible and affordable option.
West researches cheatgrass and fire regimes in Wyoming and Colorado and was a mentor for this DEVELOP project.
Cheatgrass is not suitable forage for mule deer, and its encroachment threatens their native habitat due to changes in fire regimes. Image Credit: Greg Westfall
ÛÏThe map products produced by the team will save land managers valuable time and resources that would be required if the same maps were created using traditional field surveys,Û she explained. ÛÏFurthermore, indirect long-term benefits of this project include improved forage for mule deer and elk in the study area and enhanced recreational forest use.Û
Haynes touched on the importance of using cost effective options for cheatgrass mitigation as she broke down the expenses associated with aerial herbicide application.
ÛÏAlthough it’s very efficient and per acre it’s pretty cheap to spray aerially, it’s still a big cost,Û she said. ÛÏAnd we are talking about a lot of acres.Û
The price of helicopter spraying and herbicide combined comes in around $31.60 per acre; in a 100,000-acre study area this equates to an expensive endeavor. Haynes says the MBNF will target the most heavily infested areas using the maps DEVELOP provided without having to spend great effort on the ground. This method will allow for the treatment of a maximum amount of affected areas using available funds.
The MBNF is working with state and local partners, including several wildlife conservation groups, to fund a $100,000 project to treat 3,166 acres of cheatgrass in the nearby Squirrel Creek burn area. A newly completed Environmental Impact Statement (EIS) for the MBNF analyzed the effects of aerially spraying and other weed treatments across the forest, including these fires and similar areas. While $100,000 may seem like a lot of money, according to the EIS the costs of not mitigating cheatgrass in these areas would likely be much higher. The EIS for the treatment of invasive plant species highlights potential socio-economic effects of not treating infested areas, which include: reduced land values, increased livestock related expenses due to decreased forage availability, elimination of sagebrush dependent species such as sage grouse, and reduction of huntable populations of mule deer, pronghorn, and bighorn sheep.
This decrease of natural goods and services would negatively impact the well-being and economic stability of citizens in the region. The EIS determined the cumulative socio-economic impacts of aerial herbicide application would be less costly than other alternatives, including the option to not treat cheatgrass in this area.
Tanks filled with water for use in wildfire mitigation sit on a hillside in Lebec, California. Invasive species such as cheatgrass, tamarisk, and eucalyptus are known to alter fire regimes in this part of the country. Image Credit: George Garrigues
National and Global Application
Cheatgrass is not only an issue for MBNF; the U.S. Forest Service (USFS) is equally concerned with the potential effects of cheatgrass outbreaks nationwide. The USFS addressed the severity of its concern regarding cheatgrass and other invasive weeds in their 2015 budget.
According to the report, 4,005 species of forest plants and animals are at risk of extinction, and while non-specific habitat degradation is the primary threat, 49 percent of these imperiled species are threatened by non-native species invasion.
The use of remote sensing and Geographic Information Systems (GIS) in mitigating and monitoring invasive species has the potential to preserve time, money, ranchland, recreational opportunities, and healthy ecosystems not only in the MBNF, but in areas impacted by invasive species worldwide.
The availability of this type of data analysis may be an important part of the solution needed to allow sites like the Arapaho fire and other areas affected by invasive weeds time to heal and return to their natural rhythms. These rhythms are valuable not only to the plants and animals that depend on them, but also to the people who enjoy and utilize the natural resources they provide.
Chandra Fowler is an alumna of the NASA DEVELOP program and lives in Durango, Colorado. Fowler’s research has focused on remote sensing, eco-forecasting, species modeling, and endangered/endemic species distribution. More of her writing can be found at her website travelingforchange.org.