Key points are not available for this paper at this time.
On the 19th of July 2023, a severe thunderstorm passed over Croatia, causing remarkable wind damage in the forests along a ~300 km long track of the storm from the countrys western border, at first hitting Mount Medvednica and nearby capital Zagreb, passing along the Sava river lowlands, all the way to Croatias eastern border. One of the most affected and largest contiguous areas struck by the storm was the Spava pedunculate oak forest in the eastern part of Croatia. However, many smaller areas were heavily affected across the country, too. Surveying the affected areas in the field might be a longer process, due to the need for cleaning after the considerable amount of debris and remaining dead wood which obstruct passage. Our aim was to support this survey of the damage by remote sensing measurements. Due to the fact, that the affected area is large, with a country-scale, the uniform detection and assessment of the damage can be made basically only with space-borne remote sensing. While the spatially explicit detection requires datasets with fine spatial resolution, the statistical methods rely on longer time series.In our study, to fulfil this need, we used the Harmonized Landsat Sentinel (HLS) v2.0 dataset with 30-meter spatial resolution to detect the damaged areas and with that the exact track of the storm along a 300 km long path and assess the magnitude of the caused damage. The main advantage of this dataset is its fine temporal resolution, which facilitated accurate temporal detection of the forests with damage related changes in their phenology. The damaged areas were identified based on the drop of vegetation indices (NDVI and EVI) after the storm, while to the damage assessment we used data for the whole joint Landsat Sentinel era (20162023) as well. As validation, the daily data of the commercial Planet satellites with 3-meter resolution were utilized. Beyond the remote sensing data, forestry data was also used as information on the species, age, wood volume stocks, and management operations (thinning or harvesting). Our results showed that the free HLS dataset is quite appropriate for the detection of storm damage in a wider area, but in the assessment of the damage the data from the existing forestry management plans and/or surveys are highly beneficial. Also, the detected areas are under the effects of several other factors as well, such the spreading invasive oak lace bug, making the detection more challenging.Funding: The research has been supported by the Hungarian Scientific Research Fund (OTKA FK-146600), by the Croatian Science Foundation project MODFLUX (HRZZ IP-2019-04-6325), and by the TKP2021-NVA-29 project of the Hungarian National Research, Development and Innovation Fund. Project no. 993788 has been implemented with the support provided by the Ministry of Culture and Innovation of Hungary from the National Research, Development and Innovation Fund, financed under the KDP-2020 funding scheme.
Kern et al. (Sat,) studied this question.