Fjällbjörkmätarlarvers påverkan på fjällbjörkskog : en fjärranalysstudie om markvegetationsförhållanden i subalpin miljö

Sammanfattning: Since the mid-1950s, studies have frequently been conducted on couplings between geometrid moth outbreaks and mountain birch, though not nearly as much focus has been on connections between geometrid moth outbreaks and ground cover. Previous studies have shown that major outbreaks of Epirrita autumnata occur in 9 – 10-year cycles (Karlsson et al., 2004). Sometimes also the Operophtera brumata are present in outbreaks in Abisko (Bylund, oral comment). Not until the larvae have finished feeding on the mountain birch, do they move on to the ground cover (Bylund, oral comment). Historically notorious outbreaks of geometrid moths occurred 1954 – 1955 (the most comprehensive outbreak on record), 1963, 1973 – 1974, 1984, 2003 – 2004 (with some registered repercussions in 2005) and 2012 – 2014. The study was aimed at finding methods to describe how geometrid moth larvae affect the mountain birch forest as an ecosystem. Like many previous studies, the study treated the regions around Abisko, Vadvetjåkka and Kopparåsen/Katterjåkk (northern Sweden), which all lie in the subalpine mountain birch belt. In accordance with the Swedish Environmental Protection Agency (Swe. Naturvårdsverket), mountain birch forests were defined as regions where the mountain birch constitutes at least 50% of the surface (Naturvårdsverket, 2012). The study made use of a combination of (i) low-resolution satellite data of vegetation indices (NDVI provided by MODIS), (ii) photographs (taken from both the air and the ground level) and (iii) interviews. The interviewed individuals were people which possess highly relevant information and which together contributed with a wide knowledge of moth outbreaks in the birch forests around Abisko. The study separated spring season NDVI values from summer (high) season NDVI values. For the high season, seasonal average NDVI values were calculated for each year by taking the mean of the four values provided by MODIS for every summer. In contrast to the treatment of the high season values, the annual spring values were not mean values, but instead the highest detected value of the spring season (defined as pre-mid-June). The physical regions were subdivided into two categories – healthy and prosperous forest in 2014 and damaged forest 2014/2015. According to the results of the study, there was a set-back in photosynthetic activity in the years 2011, 2012 and 2013. There was a significant (t-test) difference in photosynthetic activity between the healthy and prosperous forests and the damaged forests. An observation revealed that the prosperous and the damaged forests also showed differences related to topography (slope and aspect) and ground vegetation species. Based on the results of the study, three conclusions can be drawn: 1. Vegetation index from low-resolution satellite data can be used to investigate vegetation conditions in mountain birch forests. According to the results of this study, there was a decline in photosynthetic activity over the years of 2011, 2012 and 2013. The areas with defoliated and non-defoliated trees differ significantly (t-test) with respect to photosynthetic activity. 2. There may be a connection between ground cover photosynthesis and insect defoliation in forest ecosystems. Defoliated and non-defoliated forests also appeared to differ in other aspects, such as topography (gradient and aspect) as well as species composition in the field layer. The results of this study indicate that these differences could be used to identify areas that are more prone to be defoliated in the future. To ensure this, however, a larger study needs to be carried out. 3. Testimonies of those interviewed provide information about people’s experiences and corroborate the findings of remote sensing.