SAVANNA LAB

The whole world is a savanna; forests and grasslands are just special cases.

Dryland ecology

We study how drylands plant communities function and interact with climate, soils, fire and herbivory: what controls woody plant populations in temperate and tropical savannas, and which processes (relating to climate, soils, fire, invertebrate and vertebrate herbivores) affect plant populations, community structure and ecosystem function at meter to landscape scales? Research includes field and greenhouse studies of plant demographic processes underlying shifts in community structure, plot-scale investigation of plant community dynamics in response to climate, fire and herbivory, and research into the role of termite mounds in landscape scale hydrology and function. We work at field sites at the Jornada long-term ecological research (JRN-LTER) site near Las Cruces, in East, West and Southern Africa and occasionally at other sites around the world.

Related publications:

Lab members involved: Julius Anchang, Wenjie Ji, Qiuyan Yu, Wade Ross, Caroline Toth, Robert Wojcikiewicz, Brianna Lind, Lara Prihodko, Niall Hanan

Savanna remote sensing

Our remote sensing activities include field based UAV, airborne and satellite remote sensing, using a variety of optical, microwave and lidar approaches at a range of scales from cm-scale UAV retrievals of canopy structure at intensive field sites, to km-scale analysis at regional and continental scales. Our research has focused on remote sensing of tree-grass (woody-herbaceous) systems, with particular emphasis on partitioning of woody and herbaceous vegetation components (e.g. woody and herbaceous leaf area index), and estimation of woody canopy cover, height and biomass.

Related publications:

Lab members involved: Julius Anchang, Wenjie Ji, Qiuyan Yu, Kaboro Samasse, Wade Ross, Robert Wojcikiewicz, Brianna Lind, Lara Prihodko, Niall Hanan

Carbon and water cycle

Our research includes analysis of carbon stocks (biomass) and fluxes in dryland and savanna vegetation, including short-term (daily and seasonal) and longer-term (annual and decadal) changes associated with physiology, demographics, land use and land cover change, shrub encroachment and woody population dynamics. These themes are approached using a combination of greenhouse and field observations and experiments, remote sensing and modeling. Our water-related research includes investigation of the role of rainfall, biotic and edaphic process in controlling evapotranspiration and energy balance, vegetation dynamics and surface water for livestock, wildlife and pastoral communities.

Related publications:

Lab members involved: Qiuyan Yu, Wade Ross, Lara Prihodko, Niall Hanan

Vegetation, ecosystem and land surface modeling

Our modeling activities include use and analysis of low dimensional ecological models describing vegetation community dynamics as impacted by climate, plant competition, fire and herbivory; 2-dimensional dynamic vegetation (tree-grass) models; and more complex 2- and 3-dimensional land surface-atmosphere exchange and models. The common theme of our modeling approaches is to inform our understanding of the ecology and function of dryland and savanna systems.

Related publications:

Lab members involved: Wenjie Ji, Qiuyan Yu, Wade Ross, Brianna Lind, Lara Prihodko, Niall Hanan

Shrub encroachment and shrub demographics

Shrub encroachment has been identified as a common phenomenon in southwestern US drylands and globally. However, the rates and underlying causes, whether related to climate change, land use and disturbance, or other processes are not always clear (or the same). We are investigating both the rates and causes of shrub encroachment at the Jornada Basin LTER site and globally, using a combination of field and greenhouse experiments, regional-global scale observations via meta-analysis and remote sensing, to better understand where and why shrub encroachment occurs.

Related publications:

Lab members involved: Wenjie Ji, Caroline Toth, Robert Wojcikiewicz, Niall Hanan

Data analytics

Field experiments, remote sensing, geospatial analysis and modeling generate large and complex datasets. We are use high performance and cloud based computing resources to process and analyze large data associated with large area and high resolution remote sensing, using machine learning, simulation and statistical model. We are particularly interested in interpretable machine learning focused on accurate prediction with enhanced understanding of underlying processes and ecological insight.

Related publications:

Lab members involved: Julius Anchang, Wenjie Ji, Qiuyan Yu, Wade Ross, Lara Prihodko, Niall Hanan