Period: 01.10.2023 – 30.09.2026

Funding SOURCE: ARIS

REFERENCE NR.: J7-50095

PROJECT PROGRAMME: basic research project

COORDINATORS AND CONTACTS: assoc. prof. dr. Niko Lukač

Project website: gemma.feri.um.si/sampa/

LINKEDIN: /

Spatiotemporal algorithms for Microclimatic Parameters Assesment

Abstract:

In recent years, extreme microclimate patterns at various locations around the world have caused significant environmental and economic damage. As the current climate policy of the European Union remains inadequate for achieving the Paris Agreement’s temperature increase limit, the European Commission has introduced the European Green Deal (EGD) with the aim of making the EU climate-neutral by 2050. As part of the EGD, the Commission has proposed a reduction in greenhouse gas emissions by 55% by 2030, as outlined in the 2030 climate and energy framework. For the period 2021–2027, more than 150 billion EUR have been allocated to mitigate the socio-economic impacts of transitioning to a climate-neutral economy. It is therefore essential to understand the patterns of climate change on both local and global scales and to provide appropriate prescriptive and predictive computer analytics to enhance mitigation measures. Simultaneously, Earth Observation (EO) data acquisition, using remote sensing and field measurements, has increased more than tenfold in recent years. This provides new opportunities to improve decision-making and monitoring of microclimatic parameters. Environmental simulations and machine learning using EO data are currently among the most promising solutions for more accurately assessing more complex environmental phenomena in spatial and temporal dimensions.

With the proposed basic interdisciplinary research project Spatio-temporal algorithms for microclimatic parameters assessment (SAMPA), we will effectively address these challenges by structuring extensive, high-resolution spatio-temporal EO data into an appropriate 4D structure, which will be used as a multi-resolution input for newly developed parallelized environmental simulation algorithms. By combining data from multiple environmental simulations with structured EO data, it will be possible to assess microclimate environmental parameters with sufficient spatial accuracy over a larger geographic area, both spatially and temporally (4D). We anticipate the ability to structurally link EO data from different sources in spatial and temporal dimensions for a given location. We will validate the algorithms with three pilot tests (Pilot 1: ground temperature assessment, Pilot 2: snow cover change assessment, and Pilot 3: air quality assessment), which will be executed using general-purpose computing on graphics processing units within a high-performance computing framework. We expect improvements in accuracy in both spatial and temporal dimensions compared to the current state of the art, as well as a significant acceleration of environmental simulation algorithms compared to non-parallelized implementations. The results of each pilot will be presented on state-of-the-art geographic information system infrastructure.