Decoupling the effect of the forest pest damage from the effects of meteorology using space-borne remote sensing and modelling

The EU’s climate change mitigation plans of 55% reduction in greenhouse gas emission by 2030 and reaching climate-neutrality by 2050 rely significantly on maintaining and increasing the carbon sink in European forests. In addition to direct consequences of climate change and ageing forests, this sink is becoming threatened by the new invasive forest pests which can decrease forest productivity. The Oak lace bug (Corythucha arcuata, Say 1832), native to North America, is a new invasive species rapidly spreading since 2012 from the east to the west of Europe. The oak lace bug (OLB) after establishment in an area shows no signs of retreating and negatively affects the tree photosynthetic capacity by feeding on leaf sap. The consequences of such new and persistent pest, which are not imminently life-threatening to trees but are long-lasting, have yet to be determined.

In our study, we used remotely sensed MODIS NDVI (MOD09Q1), gridded meteorological data (FORESEE), soil water content (ERA5 Land), available national forest management and land cover data to develop methods for detecting the presence and the assessment of the impact of the OLB. The study was focused on the modelling tools to decouple the effects caused by the environmental variables from the pest damage on the measured NDVI. To this different NDVI models were created based on the Least Absolute Shrinkage and Selection Operator (LASSO) technique and the most influential periods, to support accurate forest pest detection. We investigated forests containing oak trees in the transboundary area of Hungary and Croatia. The results show that the LASSO technique is a promising tool in NDVI modelling using meteorological and environmental data. The performance of the models based on the Most Influential Periods (MIP) of the different variables showed just slightly worse results, although their application is more intuitive. In the case of the OLB, the damage assessment results with the LASSO and MIP methods showed that the pest-caused NDVI decrease in pure oak stands during the late August to early September period can be as much as -14.5% and -15.6%, respectively.

 

Asknowledgments:

The research has been supported by the Croatian Science Foundation project MODFLUX (HRZZ IP-2019-04-6325), by the Hungarian Scientific Research Fund (OTKA FK-128709) and by the János Bolyai Research Scholarship of the Hungarian Academy of Sciences.