Desert Sensing – Characterizing recent surface dynamics in arid regions through high-performance data analytics of multi-sensor Earth Observation archives and in situ records

Climate change continues to impact diverse ecosystems. Drylands stand out as particularly vulnerable environments, as they are highly responsive to key indicators of change. The sensitivity and response time of these regions remain largely unknown, underscoring the need for a deeper understanding of their systems.

Arid regions are considered optimal for Earth Observation based research, primarily due to factors such as minimal anthropogenic disturbance, sparse vegetation cover, and low cloud coverage. These attributes make drylands advantageous for studying and monitoring the impact of climate change, providing valuable insights into these vulnerable ecosystems.

Southern Mongolia stands out as an especially well-suited study area to test novel approaches and to detect land surface changes over both space and time. The basin of Orog Nuur was selected in this study to observe long-term environmental changes, building on significant prior studies conducted around the drainage basin.

Our approach emphasizes the utilization of state-of-the-art earth observation technology to unveil the dynamics of desert ecosystems. This involves cloud-based processing, such as Google Earth Engine and the German High Performance Data Analytics (HPDA) platform “terrabyte”. Throughout the project, we will apply various multispectral and active SAR techniques spanning 50 years to monitor geomorphological processes, ecosystem changes and ongoing surface dynamics linked to climate change indicators. Some of important pillars of the long-term time series analysis can be listed as greening and precipitation events, lake level dynamics, dune movement rates, mapping of sedimentological, geomorphological provinces and aeolian coverage, in order to understand frequency-magnitude relationships.

The findings will be supported by a series of fieldworks covered by UAS campaigns and auxiliary ground-truth sensors, ensuring the accuracy of our estimations by in-situ measurements. Based on the derived surface characteristics, various ecosystems will be defined, and a high-level ecosystem integrity model will be developed. Ultimately, our model aims to represent the intactness, functioning and structure of the different ecosystems within arid regions. Additionally, due to our high temporal study concept, the model will serve as the base for quantifiable measurements of the responsiveness and adaptiveness of the ecosystems.

Having a model for ecosystem intactness not only help to preserve fragile ecosystems but also strengthens the resilience and adaptive capacity of communities. Furthermore, the transferability of our framework to other drylands may also lead to a comprehensive understanding of the arid characteristics.

Keywords: Earth Observation, arid regions, dryland, remote sensing, climate change, impact, geomorphological process, ecological modelling, land surface dynamics