Occurrences of insect outbreaks in Sweden in relation to climatic parameters since 1850

anglais. This study was carried out in order to identify the main insect species responsible for forest damages (pests) in Sweden, and to explore the relationships between insect outbreaks and environmental variables (temperature, precipitation, and availability of insect host trees, including storm damaged trees). Special attention was given to changes in management practices through time, and their consequences for pest outbreaks. The occurrences of insect outbreaks were analyzed for the southern, central and northern Sweden during two periods, 1850-1950 and 1961-2014. A Principal Component Analysis was conducted for each of the three regions in order to assess which insect families and insect species caused the main forest damage. The ratio between broadleaved and coniferous trees was calculated to highlight the type of forests mostly attacked by insect pests. A Variation Partitioning Analysis was carried out to study the influence of the climatic variables (temperature and precipitation) and the volume of storm felled trees on the occurrences of insect species responsible for outbreaks. During the first period, insect outbreaks increased markedly after 1911, and remained at a high level throughout the period. The Variation Partitioning Analysis showed that this trend might have been strongly influenced by climate and storm events. However, for the second period of time an association between these variables and the insect species responsible for the main forest damage could not be distinguished, and a clear decrease in the occurrences of insect outbreaks could be noticed in comparison with the first period. Other studies clearly showed a relation between outbreaks and climate also for the second period. The inconsistency in this study between the two time periods is attributed to the different datasets used and to other processes, such as changes in forest management practice, the use of insecticides and other countermeasures.

anglais. The objectives of this study were to observe the tendency of insects for a type of tree in order to calculate the ratio of attacked coniferous versus attacked broadleaved trees, to identify the main five insect families responsible for outbreaks and the most important insect species responsible for outbreaks (for each southern, central and northern Sweden, and for two periods of time: 1850-1950 and 1961-2014), and to observe how climatic variables (i.e. temperature, precipitation) and the available storm damaged trees have influenced the pests behavior over time. Sweden is a country covered mostly by forests (70%). By having large areas with forests, Sweden faces insect outbreaks to a big extent. Insects that cause damages to trees (pests) pose as serious threats to the tree’s health, and in some cases they cause tree mortality. By damaging the trees they induce large economic losses. It is therefore important to study insect outbreaks in order to understand the mechanism that leads to their outbreaks and try to take precise countermeasures against these pests. The software used displayed the results as arrows (insect species, insect families and climatic variables) and the years as circles. This analysis showed interesting results. For insect families, as well as for insect species different results were displayed for the two periods of time. For the first period (i.e. 1850-1950) the three regions of Sweden showed some resemblances, and some of the families and species were present in all three regions as important ones. There were only one or maximum two families and insect species that were the same between the two periods of time. Regarding the influence of climatic parameters on the behavior of the main insects responsible for outbreaks the analysis displayed interesting results. Throughout the first period (i.e. 1850-1950) insects were influenced to a big extent by the above variables, whereas throughout the second period (i.e. 1961-2014) the main insect species responsible for outbreaks showed no correlation with the above variables (i.e. temperature, precipitation and availability of storm felled trees). However it was difficult to make a direct comparison between the two periods due to the different datasets used in this analysis.