Author: Fabian Loew

LaVaCCA Introductory GIS / R course in Almaty, Kazakhstan

After the kick-off meeting of the Volkswagen Foundation funded LaVaCCA project in Almaty (see post), an intensive R/GIS training was held in the GIS lab of the Al ́Farabi Kazakh National University, organized and led by staffers from the Department of Remote Sensing, Würzburg University. Young scientific staffers, post-docs and PhD students as well as invited guests from SIC-ICWC, Al´Farabi University and CAREC joined the training course from February 12th to 17th 2015. The focus of the course was on methods and techniques for the application of satellite remote sensing in agricultural management, including the use of the statistical and open source programming language R:   Introduction to the statistical software R Spatial statistics in R Working with vector data in R Working with raster data in R Visualizing and manipulating satellite data in R This course served as a preparatory course for another, intensive training to be held in Urgench, Uzbekistan, in summer 2015. The feedback from the participants for the first course in Almaty was very positive and participants were given the opportunity to propose the topics and schedule of the second course in Urgench, Uzbekistan in summer 2015. More information on the LaVaCCA project can be found here. Information on the R project for statistical computing, including download links can be found here....

Read More

Satellite based locust habitat monitoring in the Aral Sea basin

The Central Asia research team at the Department of Remote Sensing, Würzburg University, recently started investigating potential risk areas for infestation of the Asian Migratory locust, Locusta migratoria migratoria L., in the Aral Sea basin. Asian Migratory locust is a serious crop pest in the Amudarya River delta near the Aral Sea in Uzbekistan, Central Asia. Its economic importance increased after desiccation of the Aral Sea. Vast areas of the former sea bottom became covered with stands of common reeds Phragmites australis, which are well-known as the main breeding locust habitat in this region.  In order to enable efficient locust monitoring, accurate information about the spatial distribution of reeds is essential. Traditional ground-based locust monitoring is hardly possible in most areas of the delta, particularly on the former sea bottom. Due to its high revisit frequency, adequate spectral resolution and synoptic coverage, data from the MODIS satellite sensor aboard the Terra and Aqua platforms can provide the required information about spatio-temporal reed distribution in the Amudarya delta. Additional efforts are recently spent on extending the methodology to provide early detection of potential breeding habitats within the growing season, and to analyze the correlation of water inflow to the delta and the spatial extent of locust breeding habitats. The results are instrumental for predicting potential locust outbreaks and developing better targeted management plans. First results of our research will be...

Read More

LaVaCCA Project kick-off meeting in Almaty, Kazakhstan

On 09/10 February 2015 the LaVaCCA project team from Würzburg met with the project consortium at the Al´ Farabi National Kazakh University, Kazakhstan. The initial project meeting focused on the presentation of the working packages and important organizational issues. The LaVaCCA project addresses the identification of hotspots of decreasing land production and gaining knowledge about the drivers of change in land production and LD by analysing socio-economic and ecological indicators. Economic assessments of land use options under given ecological conditions will be elaborated. A strong methodological focus is set on remote sensing, geographical information systems (GIS), indicator systems, land use modelling, and economic optimization. The generated information will be bundled and presented as a tool of discussion support for politicians and decision makers in their efforts to increase food security and combat environmental degradation in the irrigated areas of CA. After the very successful meeting, the project partners from Germany, Uzbekistan (SIC-ICWC, KRASS) and Kazakhstan (Al´ Farabi University) started planning the extensive field surveys in the lowlands of the Amudarya and Syrdarya river deltas, which are the only rivers flowing into the Aral Sea. Methodological focus is set on measuring biophysical parameters, e.g. fPAR and soil salinity for remote sensing based assessments of crop yields and salinity, respectively. Also, interviews with farmers and socioeconomic assessments of alternative land use options on degraded or abandoned agricultural lands are recently being...

Read More

New project on agricultural production in Central Asia

In January 2015, a new project on Assessing Land Value Changes and Developing a Discussion-Support-Tool for Improved Land Use Planning in the Irrigated Lowlands of Central Asia (LaVaCCA), funded by the Volkswagen Foundation was started by the Central Asia research group at the Department of Remote Sensing in Würzburg. Summary: Immense losses of land productivity have been observed on eight million hectares of irrigated agricultural land in Central Asia (CA) during the past decades. Especially the irrigated lowlands of the Amu Darya and Syr Darya Rivers are affected by land degradation (LD) problems. Despite first attempts to derive different...

Read More

New publication on classification uncertainty

A new paper published recently in ISPRS presents our ongoing research on classification uncertainty in the context of multi-temporal object-based classification, based on support vector machines (SVM). It adds a contribution to the question of which factors influence the spatial distribution of classification uncertainty in land use maps. Agricultural management increasingly uses crop maps based on classification of remotely sensed data. However, classification errors can translate to errors in model outputs, for instance agricultural production monitoring (yield, water demand) or crop acreage calculation. Hence, knowledge on the spatial variability of the classier performance is important information for the user. But this is not provided by traditional assessments of accuracy, which are based on the confusion matrix. In this study, classification uncertainty was analyzed, based on the support vector machines (SVM) algorithm. SVM was applied to multi-spectral time series data of RapidEye from different agricultural landscapes and years. Entropy was calculated as a measure of classification uncertainty, based on the per-object class membership estimations from the SVM algorithm. Permuting all possible combinations of available images allowed investigating the impact of the image acquisition frequency and timing, respectively, on the classification uncertainty. Results show that multi-temporal datasets decrease classification uncertainty for different crops compared to single data sets, but there was no “one-image-combination-fits-all” solution. The number and acquisition timing of the images, for which a decrease in uncertainty could be realized,...

Read More

news blog by

the Remote Sensing Department
at the University of Würzburg
Institute of Geography and Geology
Oswald Külpe Weg 86
97074 Würzburg

University Webpage

Recent Tweets