Vegetation Cover Assessments and Management Applications Using Multiscale Remote Sensing Across the Santa Rita Experimental Range

Remote sensing (RS) and Geographic Information Systems (GISs) provide significant opportunities for monitoring and managing natural resources across various temporal, spectral, and spatial resolutions. RS and GIS are increasingly used to collect and analyze data, monitor various aspects of the environment, and are being leveraged in decision-making throughout the fields of conservation and range-management. RS and GIS are used to compare changes in land cover over time, and track landscape health by measuring vegetation green-up and dieback, fragmentation of landscapes, and human encroachment. There is a critical need for natural resource managers to understand the expanding capabilities of image sources, analysis techniques, and on-site validation methods. As a result, understanding how to work with RS and GIS and applying the technology to existing management methods in order to understand state of lands is becoming increasingly necessary. In this 8-year research study, I began by understanding (by way of conversations, participating in field and site visits, and reviewing existing literature) the techniques current natural resource managers use to manage lands they are responsible for. I then explored several RS and GIS novel methods developed to measure the state of lands managed. I further compared and contrasted field, RS and GIS methods, and various RS and GIS applications used to measure vegetative cover over a period of time (an important environmental indicator). My research has corroborated the importance of leveraging RS and GIS technologies to aid in vegetation cover assessments and inform natural resource management activities. RS and GIS data, analysis and techniques to inform state and transition models (STMs) are presented in Appendix A. We found that remote sensing products mapped vegetative states that closely coincided with fieldwork results, including areas that were missed during the field mapping and areas that were classified as another vegetative state in the remote sensing analysis. The fine spatial resolution aerial and drone imagery, however was not able to detect if grass was native or non-native (for future study). Remote sensing is able to cover larger areas with an encompassing view - areas that are often missed in fieldwork. Comparison of field to RS and GIS methods are explored in Appendix B. Knowing the extent to which these field and remotely sensed perspectives might differ is important to land managers interpreting vegetation dynamics. We found that aerial and satellite data were strongly correlated with bare ground and grass with a less strong correlation occurring for woody cover. Both the aerial and satellite data are positively correlated with transect woody cover measurements on < 1.21 hectare sites, but weakly correlated or not at all for > 60.70 hectare sites. Interpreting ground cover, for example, tree and shrub cover among the different methods is significant for rangeland ecosystems. The importance of validating the ability of Landsat and Sentinel sensors to detect the pattern and extent of shrub cover in an arid landscape when grasses were senescent and shrubs were still green, leveraging a 15 cm multispectral 2023 aerial image are explored in Appendix C. We determined that both Landsat and Sentinel satellite data were moderately correlated to the aerial image. Further analysis showed that Sentinel and Landsat shrub cover estimates were also moderately correlated to shrub percentages from The Rangeland Analysis Platform (RAP) application. Users of these tools, data, and applications should be aware of the constraints and limitations to quantify shrub cover.