Linking Plot and Landscape Level Phenology

Plant phenology has become popularize since the 1990s to help understand the effects of climate change on vegetation, it is a way to study the patterns of leaf-out, flowering, and senescing of the life-cycle of the plant. The timing from start of season (SOS) to end-of-season (EOS) of plant leaf-out development has important implication on ecosystem productivity and carbon cycling across ecological scales from individual trees to whole regions. The need for plant phenology studies in dryland are highly anticipated, since they make up 40% of the terrestrial landscape and recently have shown the play an important role in the trend and variability of carbon. The main drivers of leaf-out phenology in drylands are complex to understand due to the high spatial heterogeneity from plot to regional scales and variable precipitation patterns, increasing the need for monitoring tools at different scales. Multiscale, long-term phenological monitoring allows for the advancement of understanding these drivers and their mechanism. Ground level phenology has its shortcomings in drylands, not only due to the harsh environments and the effort to deploy individuals at the plot level in various regions thus enabling spatial distribution inconsistencies, but also it involves different observers and methods adding more inconsistencies in the quality of the data. Several national agencies like the National Phenology Network has organized to improve on standardizing definitions but there are still varied definitions for individuals stages that are adopted among different researchers. Remote sensing phenology, like UAV and satellites, have gained importance in capturing the leaf-out times at different spatial and temporal scales. The purpose of this study is to discover new ways of measuring leaf-out transition cycle patterns through the use of remote sensing like UAV by developing classification maps of the study site and furthering the steps to scale plot and landscape. Plot level leaf-out patterns of creosote and honey mesquite were recorded from 2010-2019 with different observer taking data over the years and were explored against the temperature and precipitation. The results showed that it was difficult to make a clear connection between the climate data and vegetation, especially for the creosote due to high significant attribute of observer variability. However, temperature for both species showed to be an important factor, but more analysis to see this clearly is still needed. The classification map created through UAV/RGB sensor had a significant accuracy, enabling further discovery to leverage this technique to accurately represent drylands within the broader Earth System in understanding leaf-out development