Hanna gave an interview in the Podcast of the "Zentrum verantwortungsbewusste Digitalisierung" (ZEVEDI) in March 2023 about AI in environmental remote sensing. Listen to the Interview here.

Our master student Candy Fahrenholz was working in the Kuzikus Wildlife Reserve in Namibia as part of her master thesis. One focus of her research was to survey the small-scale landscape structures using drone-based remote sensing. With the help of this high-resolution image data, the research question "Which environmental parameters affect nest tree selection of endangered vulture species in Namibia?" will be answered.

Marvin Ludwig successfully defended his dissertation on 10.02.2023 and was awarded a doctorate in natural sciences. The title of the thesis is: „Remote sensing and machine learning for multi-scale ecosystem monitoring”. Congratulations from your working group!

Global-scale maps of the environment are an important source of information for researchers and decision makers. Often, these maps are created by training machine learning algorithms on field-sampled reference data using remote sensing information as predictors. Since field samples are often sparse and clustered in geographic space, model prediction requires a transfer of the trained model to regions where no reference data are available. However, recent studies question the feasibility of predictions far beyond the location of training data. Here, we propose a novel workflow for spatial predictive mapping that leverages recent developments in this field and combines them in innovative ways with the aim of improved model transferability and performance assessment. We demonstrate, evaluate and discuss the workflow with data from recently published global environmental maps. The publication can be found here: Ludwig, M., Moreno-Martinez, A., Hölzel, N., Pebesma, E., Meyer, H. (2023): Assessing and improving the transferability of current global spatial prediction models. Global Ecology and Biogeography.

In the last 15 month, we measured the soil temperature and the soil moisture at 250 different locations in our 20 x 20 km large study area in the Fichtel Mountains. At each location the measurement duration was one week or one month. Thus, we now have a dataset with measurements at different locations (forest sites, meadows, and fields) under a variety of meteorological conditions. Now we want to use the data set to model the soil temperature and the soil moisture in the study area continuously in 4D, i.e., for each location, the entire time span, and every soil depth.

Our second project meeting of the Carbon-4D project took place in the Fichtelgebirge on October 7. Together with the Soil Ecology of the University of Bayreuth, the current status of the project and initial results were discussed. The measurements of soil temperature and moisture have been running since September 2021 and will continue until December this year. Further information on the project can be found on the project page [en].

We had a great week in the Harz mountains (12.9.22-16.09.22) where students from our institute came together with students from the university of Göttingen, HAWK Göttingen as well as from the University of Würzburg. We learned about current challenges in forestry in the Harz region and how we can support a monitoring of the forests with remote sensing techniques. Especially the interdisciplinarity of the group (students of landscape ecology, forestry, forest science, remote sensing) was a great experience. During Walk&Talk through the National Park, we had the opportunity to exchange ideas between the groups and then to combine the different areas in a joint project work. We are already looking forward to the next joint event next year!

This year's OpenGeoHub Summer School took place from 28.8.-3.9.2022 in Siegburg. As in previous years, there was also a workshop from us, this time on the topic of „Machine learning-based maps of the environment: challenges of extrapolation and overfitting“. The course material is available on Github. The recordings of the lecture and exercise are available. The processed versions will be linked later on our course-website.
 

Our recently published paper on "Machine learning-based global maps of ecological variables and the challenge of assessing them" [en] was picked up by the WWU press office in an article. The article can be read in the online version of the July 2022 issue [de].
 

As part of the study project Amtsvenn the first UAS flight campaign was successfully conducted on the 3rd of March 2022. High-resolution multispectral and very high-resolution RGB imagery was collected with the WingtraOne. The campaign was supervised by the “Biologische Station Zwillbrock e.V.” to survey the low disturbance of the UAS on wildlife and no disturbances were observed.
The study project investigates how UAS imagery can be used to detect and spatially model the humidity and vitality of bog vegetation. For ground truth, humidity and temperature are measured in a measurement setup developed by the students, representing different gradients (elevation, vegetation, degradation) of the bog.
This study projects follows the ongoing research in the Vechta bogs and kicks off the in April 2022 starting “Reversal Project” [en].
There are many research project, bachelor and master thesis topics for students interested in remote sensing and spatial modelling in bog ecosystems. If interested, please contact Jan Lehmann [en], Hanna Meyer [en] and Laura Giese [en].

What to do when the spatial or temporal resolution of satellite data is insufficient for the research topic? Our study shows how the strengths of two satellite sensors in terms of spatial and temporal resolution can be combined in a data-driven approach. As a result, from 1999 on, we can generate Land Surface Temperature data with a subdaily temporal and 30m spatial resolution for the Antarctic Dry Valleys – a dataset that can be used for various research applications, such as species distribution modelling or to model further relevant environmental information.

• Lezama Valdes L-M, Katurji M, Meyer H. 2021: A Machine Learning Based Downscaling Approach to Produce High Spatio-Temporal Resolution Land Surface Temperature of the Antarctic Dry Valleys from MODIS Data [en]. Remote Sensing 13, 4673. doi: 10.3390/rs13224673