Modelling day-to-day stem diameter variation and annual growth of balsam fir ( Abies balsamea (L.) Mill.) from daily climate

Abstract

In the present context of global climate changes and the continuous development of forest management strategies based on the concept of sustainable use, it is important to develop a better understanding of the climatic factors controlling the growth of boreal forests. In this study, we report the results of a five-year field research within which day-to-day balsam fir ( Abies balsamea (L.) Mill.) stem diameter variation was measured with dendrometers and examined in relation with various daily climatic variables. A model built with data from three growing seasons that included solar radiation, relative humidity, temperature and precipitation explained 84% of the variance in day-to-day stem diameter variation from June to September. The model has approximately the same predictive capability when validated with independent daily data from two other growing seasons. The model captured both the cumulative increment associated with the irreversible growth and the high frequency variation of day-to-day fluctuations associated to changes in the stem water content. In general, rainy days during which relative humidity was high and solar radiation was low favored stem diameter expansion (growth and swelling) while stem diameter decreased during periods of low relative humidity and high solar radiation. Similar models were obtained when the June–September period was divided into two parts (June–July and August–September) to better represent the period during which most of the cumulative annual stem increment is observed (June–July). Inter-annual variation in stem growth computed from the modeled day-to-day variation in stem diameter was significantly correlated to the inter-annual variation in annual growth determined from tree core measurements over a 10 year period ( p = 0.023). The model was notably able to capture a particularly poor growing year (2006) presumably due to a short-term heat stress period. Results suggest that the inclusion of daily data in growth–climate models may contribute to improve predictions of the potential tree growth response to climate by identifying particular climatic events that may escape to a classical dendroclimatic approach.