Remote Sensing to Quantify Woodland Canopy Cover

Introduction

Image Details

The San Carlos Apache Reservation spans across 1.8 million acres in Arizona and contains many unique and rare ecosystems. These ecosystems range from deserts and grasslands that have low canopy cover of woody plants, to savannas, woodlands, and forests that have higher woody plant canopy cover. The San Carlos Apache have stewarded these lands since they arrived in the Southwest in the 10th century. The San Carlos Apache Forest Resources Program (SCAFRP) continues to steward these ecosystems for their cultural, food, and medicinal value; their importance as wildlife habitat; and as a source of resources and livelihoods for Tribal members.

According to the Traditional Ecological Knowledge (TEK) of community Elders, grasslands used to be more widespread, fires more frequent, and woody plants less abundant and larger than they are today. This knowledge is supported by tree ring data that indicate frequent low-intensity fires were important for maintaining grasslands, savannas, woodlands, and forests. The SCAFRP aims to restore the land they manage to the conditions they were in before the establishment of the reservation in 1872. They are using the TEK of Elders and scientific data from tree rings, remote sensing, and other sources to develop a more precise understanding of this pre-reservation reference condition. They aim to meet this restoration goal by actively managing grasslands and savannas for prevention and reversal of woody species encroachments, and woodlands and forests for healthy tree spacing and ecologically appropriate fire frequency. 

Limited staff, resources, and availability of quantitative data about the historical and current landscape conditions make it difficult for the SCAFRP to develop quantitative land management goals and evaluate the efficacy of their management actions. Thanks to a long-term collaboration with the United States Geological Survey (USGS) Western Geographic Science Center, the SCAFRP recently completed an analysis of the earliest available aerial photographs, taken in 1935, and remote sensing data from 2017 to help understand the extent and composition of historical grasslands and savannas. This has helped refine restoration and management goals as well as evaluate the effectiveness of past and current management strategies.

Key Issues Addressed

Across the Southwest, ecosystems face increasing pressure from a warming climate, and increasingly frequent drought. Changes in land use and disturbance patterns such as overgrazing, and reduced frequency of beneficial low-intensity fires, are putting more pressure on Southwestern ecosystems. In grasslands and savannas, these pressures are leading to an increase in canopy cover of woody plants, like juniper (Juniperus spp.). Some areas of the San Carlos Apache Reservation that were historically grasslands and savannas, with a cover of <10 trees per acre, now have over 180 trees per acre. This level of encroachment decreases available habitat for grassland-dependent wildlife, increases soil erosion, and negatively alters fire regimes. In woodlands and forests, increases in canopy cover of woody plants is creating denser forests and changing the fire regime from frequent low-intensity fires, to high-intensity stand-replacing fires.

The lack of quantitative data about the grasslands, savannas, woodlands, and forests makes it difficult to establish baseline conditions and understand the historical extent of these ecosystems. It also hinders development of quantitative goals for management and restoration. Namely, where to implement what kind of management action to meet restoration goals. 

Resources for monitoring and evaluating effectiveness of treatments are also limited. There is currently only one rangeland specialist covering the 1.8 million acres of the San Carlos Apache Reservation. Using remote sensing technology can increase capacity for assessing restoration and management outcomes and evaluating if SCAFRP’s management actions are meeting the goal of restoring historical conditions in grasslands, savannas, woodlands, and forests.

Past management and restoration has not always resulted in long-term effectiveness. Many grassland areas where juniper was removed by chaining in the 1950s and 1960s are again dominated by juniper. The SCAFRP wants to identify the most effective combination of strategies to maintain health of grasslands, savannas, woodlands, and forests in terms of both cost, and long-term effectiveness to help meet their goal of returning the land to historical conditions.

There is a long history of broken agreements and treaties between Federal government agencies and sovereign Tribal nations. The San Carlos Apache Tribe has had negative relationships with researchers who sought to extract knowledge instead of collaborate for mutual benefit. Repairing and maintaining relationships of trust is critical.

Project Goals

• Apply novel image analysis techniques to historical aerial photographs from 1935 and remote sensing data from 2017 to develop a better idea of historical conditions and change in woody plant canopy cover in grasslands and savannas. 

• Combine the results of this data analysis with the TEK of Elders in the community to develop quantitative goals for restoration activities in grasslands and evaluate the effectiveness of recent grassland restoration and management.
• Determine the most effective combination of methods to use in restoration activities moving forward based on management experience and remote sensing data.
• Maintain a relationship of trust between SCAFRP and USGS to collaboratively develop information and tools for management of grasslands, savannas, woodlands, and forests.

Project Highlights

Two Decades of Collaboration: USGS and SCAFRP have been working together to produce land management tools since 2004. Consistent communication, annual visits, and collaborative problem-solving processes have been fundamental to this partnership. 

• Unique Analysis: Researchers used a unique pixel-based approach that combined two programs, ERDAS Imagine and Adobe Photoshop, to delineate canopy cover of woodland trees based on aerial photography from 1935 and remote sensing data from 2017 (Middleton and Norman, 2021). This helped the SCAFRP develop baseline, historical conditions and quantitative goals about where to remove woody plants and what the restoration target for woody plant cover may be for a given area.
• Increase in Woody Plant Canopy Cover: Many areas that appeared as woodlands with high canopy cover of woody plants in 2017 were savannas with an open canopy structure structure
  Something temporarily or permanently constructed, built, or placed; and constructed of natural or manufactured parts including, but not limited to, a building, shed, cabin, porch, bridge, walkway, stair steps, sign, landing, platform, dock, rack, fence, telecommunication device, antennae, fish cleaning table, satellite dish/mount, or well head.
  
  Learn more about structure in 1935. Additionally, areas that were historically grasslands were encroached upon by woody species, most notably juniper.The remote sensing analysis showed that the percent canopy cover of woodland trees more than doubled (from 18.3% to 38.5%) in one study area, and increased from 28.3% to 43.3% at another. 
• Denser Juniper: Many savanna and woodland areas where canopy cover was already dominated by juniper in 1935 had a higher canopy cover of juniper in 2017. It is possible that reduced frequency of low-intensity fires contributed to this tree “infilling” in savannas and woodlands between 1935 and 2017, increasing the risk of catastrophic wildfire. 
• Less Pine: Forests dominated by ponderosa pine (Pinus ponderosa) in 1935 had less pine canopy cover in 2017 than in 1935. This reduction in pine is possibly due to the severe wildfires that have killed entire stands of ponderosa pine. 
• Juniper Branches Carry Fire: For fire to be most effective at maintaining grassland structure, SCAFRP found that it was optimal to implement prescribed burns six to nine months after juniper removal. This allowed juniper brush to be dry, but still have needles attached to the downed wood and branches which helps the fire carry across the landscape.

Lessons Learned

Image Details

The long-term partnership between the San Carlos Apache Tribe and the USGS has fostered trust and good working relationships that enabled this work to happen. Many research partnerships are short-term and end as soon as the data is collected. This is extractive, as both parties do not benefit equally from data collection, and is detrimental to sustained relationships. SCAFRP and the USGS Western Geographic Science Center have been working together on multiple projects since 2004. Instead of presenting a proposal for research to the SCAFRP staff, USGS researchers instead asked what kind of information and questions the Tribe is interested in answering and allows this to dictate the scope of work. This co-development of research questions and priorities allows the results of USGS research to be highly applicable and useful to land management on the San Carlos Apache Reservation. The Tribe reviews the data and potential publications for information they deem sensitive, and the USGS responds to their decision by removing that information from reports and publications. This helps give the Tribe control and sovereignty of the data and research conducted on Tribal land. 

The analysis of aerial photography and remote sensing data showed that the grasslands and savannas that SCAFRP has been managing to prevent woody plant encroachment have not seen an increase in woody plant canopy cover. Nearly all areas that had less than 15% canopy cover of woodland trees in 1935 increased in canopy cover except for those under active management by SCAFRP. Areas that have not been actively managed and restored have seen significant increases in woody plant canopy cover indicating that, without intervention, historical grasslands and savannas on the San Carlos Apache reservation are likely to continue shrinking as woody plant canopy cover increases. These findings confirm the TEK of Tribal Elders and supplemented them by helping managers know where to focus removal efforts and what level of woody plant canopy cover to restore. 

The analysis of images and remote sensing data confirmed the effectiveness of the combination of methods the SCAFRP has developed for grassland and savanna restoration. Starting with mechanical shredding, managers reduce juniper cover to <10 trees per acre. Managers found that using a mechanical brush shredder, followed by a small chainsaw crew to remove any trees that were missed by the shredder, was approximately 10% more effective at preventing juniper regrowth and reduced costs from ~$400/acre to $75-100/acre compared to using saw crews alone. However, cutting alone, either with chainsaws or brush shredders, saw 90+% regrowth of juniper within six months. By combining cutting with an application of undiluted glyphosate (Roundup) to cut stumps, managers reduced regrowth rate to <5%. The large size of woodchips and slash left behind from shredding operations acts as a mulch that retains soil moisture, and acts as nursery logs to protect young grass seedlings from predation as they emerge. Hand saw crews and grubbing do not result in the same beneficial mulch cover. Therefore, leaving the slash spread out over the treatment area was preferable to piling it in brush piles to burn. 

Next Steps

• Locate additional areas for restoration, using the results from the analysis of remote sensing data and historical aerial photographs, to return woodlands to more savanna-like environments. 
• Determine appropriate restoration techniques to use (tree removal, prescribed burning, reseeding, etc.) in which location by combining maps of woody plant canopy cover with information about soil, grass cover, topography, seed bank, and invasive species invasive species
  An invasive species is any plant or animal that has spread or been introduced into a new area where they are, or could, cause harm to the environment, economy, or human, animal, or plant health. Their unwelcome presence can destroy ecosystems and cost millions of dollars.
  
  Learn more about invasive species .

Funding Partners

• Land Change Science Program of the U.S. Geological Survey
• U.S. Bureau of Indian Affairs (BIA) Forest Climate Change and Geospatial Capacity grant

Resources

• Tribal Land Vegetation and Watershed Modeling - US Geological Survey 
• San Carlos Apache Tribe, Inter Tribal Council of Arizona
• Mapping Riparian Vegetation Response to Climate Change on the San Carlos Apache Reservation and Upper Gila River Watershed to Inform Restoration Priorities: 1935 to Present - SW Climate Adaptation Science Center
• Presentation by Laura Norman: USGS and the San Carlos Apache Tribe
• Middleton, B. and Norman, L. (2021) ​​”Remote Sensing Analysis to Quantify Change in Woodland Canopy Cover on the San Carlos Apache Reservation, Arizona, USA (1935 vs. 2017).” Land 10:393
• United States Geological Survey Western Geographic Science Center
• San Carlos Apache Forest Resource Program

Contacts

• Cassie Moses, SCAFRP: cmoses@forestry.scat-nsn.gov
• Bary Middleton, USGS: bmiddleton@usgs.gov

CART Lead Authors

• Ariel Léger, CCAST Coordinator, University of Arizona: arielleger@arizona.edu
• Nicholas Katz, Student Intern, University of Arizona

Suggested Citation

Léger, A., and Katz, N. (2022) “Remote Sensing to Quantify  Woodland Canopy Cover on the San Carlos Apache Reservation” CCAST. Retrieved from https://www.fws.gov/project/remote-sensing-quantify-woodland-canopy-cover.