Research

Being mobile is an essential prerequisite for participating in social life. Rising energy prices have brought the issue of mobility to the fore and raised the question of the future affordability of (car) mobility. As part of MOSAIK, we want to take an in-depth look at what mobility poverty is, which groups of people and regions are affected by it - especially in terms of work and education routes - and which measures are promising for different target groups. In a first step, a spatial analysis of regional differences in mobility poverty will be carried out. For this analysis, a new data set will be created that contains the characteristics of public transport connections (e.g. duration, transfers, number of connections during rush hour, in the evening, at weekends and at other off-peak times) between important economic/company locations and large conurbations in the Innviertel region. The analyses show where the need for commuting routes can be met particularly well or particularly badly by public transport. In addition, the data from the supplementary consumer survey conducted by Statistics Austria will be used to identify population groups affected by mobility poverty on the basis of predefined indicators and to outline their spatial distribution. The result of both analyses will be a collection of particularly vulnerable regions and groups of people. With the aid of expert interviews and interviews with affected groups of people, framework conditions and targeted solutions for those affected by mobility poverty are to be developed and then discussed with regional representatives. Using a pilot study, an existing solution approach in the Innviertel region will be evaluated as an example and its transferability discussed. The results of the project will be incorporated into an action plan to reduce mobility poverty.

Vegetation mapping is an essential component in the domain of nature and environmental protection. Traditional approaches, aligned with current guidelines, necessitate high research specifications typically fulfilled through terrain mapping efforts. Despite its efficacy, this method encounters limitations in terms of seasonality and the timely processing of extensive areas. In contrast, remote sensing-based models offer noteworthy advantages under their season-independence and rapid large-scale processing capabilities. The present initiative seeks to leverage new technologies, such as cloud computing, to augment conventional supervised remote sensing classifications by merging ecological expertise with the sophisticated capabilities of cloud computing technology. For a few years now, the “Google Earth Engine” (GEE) platform has made it possible to carry out geospatial processing data and analyze them based on a huge time satellite imagery series at large study area in combination with multivariate statistical methods. It also enables the integration of location and laser scan data as well as geospatial information systems data. Challenges and Research Requirements in the project: Integrating the processing chain into a cloud platform poses considerable challenges and necessitates extensive research for a coherent, smooth, and consistent adaptation. From the initial model selection to the subsequent post-processing phase, passing through exhaustive feature selection analysis and model evaluation, which are crucial phases of adaptation, require exhaustive research to ensure the accomplishment of expected outcomes in the project. Recognizing the complexity of this task, the Egger Natural Space Planning Company requires the expertise of a remote sensing scientist specializing in cloud computing platforms. The scientist, their expertise in the field, and their scientific input are vital for providing a perspective and conducting the needed research to identify optimal approaches that align with the project's expected outputs.

The aim of the study is a consistent estimation of a wide range of low-flow characteristics at observed and unobserved river sites in Austria. For this purpose, novel regionalisation models are to be developed and evaluated with regard to their predictive performance at unobserved sites. A nested model is proposed as the model structure, which takes into account low flow (NQ) of different time scales (year, season, month, minimum observed value) in hierarchical form, in order to obtain consistent estimates of derived mean characteristic values and extreme value statistics. The regionalisation aims at a consistent estimation of natural low flows and uses a data set of at most slightly anthropogenically influenced daily flow series with more than 40 years of observations.