Comparing statistical and mechanistic models to identify the drivers of mortality within a rear-edge beech population

Abstract

31Since several studies have been reporting an increase in the decline of forests, a major issue in ecology 32 is to better understand and predict tree mortality. The interactions between the different factors and 33 the physiological processes giving rise to tree mortality, as well as the individual between-tree 34 variability to mortality risk, still need to be identified and assessed. 35This study is based on a survey of 4323 European beeches (Fagus sylvatica L.) since 2002 in a rear-36 edge population within a natural reserve. We combined two types of approaches: (1) statistical 37 models were used to quantify the effects of competition, tree growth, size and decline on mortality 38 and (2) an ecophysiological process-based model (PBM) was used to separate out the different 39 mechanisms giving rise to temporal and inter-individual variations in mortality by simulating carbon 40 reserves, hydraulic conductance and late frosts in response to climate. 41The mortality rate at population level was associated to the combination of conductance loss, carbon 42 reserve depletion and occurence of late frosts simulated with the PBM. In the statistical models, the 43 individual probability of mortality decreased with increasing mean growth, and increased with 44 increasing crown defoliation, earliness of budburst, fungi presence and increasing competition. The 45 interaction between tree size and defoliation was significant, indicating a stronger increase in 46 mortality associated to defoliation in smaller than larger trees. Finally, the PBM predicted a higher 47 conductance loss together with a higher level of carbon reserves for trees with earlier budburst, while 48 the ability to defoliate the crown was found to limit the impact of hydraulic stress at the expense of 49 the accumulation of carbon reserves. 50We discuss the convergences and divergences obtained between statistical and process-based 51 approaches and we highlight the importance of combining them to identify the different processes 52 underlying mortality, and the factors modulating individual vulnerability to mortality. 53 54 55 56

Genetic signatures of divergent selection in European beech (Fagus sylvatica L.) are associated with the variation in temperature and precipitation across its distribution range

“…Beech putative glacial refugia and colonization routes have been identified based on very detailed pollen records and genetic population surveys with chloroplast and isozyme markers (Magri et al, 2006). Beech is known to be highly sensitive to summer droughts (Aranda et al, 2015; Knutzen et al, 2017), and, to a lesser extent, to late frosts (Kreyling et al, 2014; Petit‐Cailleux et al, 2020). Genetic variation has been investigated at various climate‐related phenological traits (Gárate‐Escamilla et al, 2019; Gauzere et al, 2020; Gömöry & Paule, 2011; Kramer et al, 2017; Vitasse et al, 2009), physiological or morphological traits (Bresson et al, 2011; Hajek et al, 2016; Wortemann et al, 2011) and performance traits (Gárate‐Escamilla et al, 2019).…”

Tree Mortality Risks Under Climate Change in Europe: Assessment of Silviculture Practices and Genetic Conservation Networks

“…The CASTANEA version we used allows PLC to be computed based on daily midday water potential and species vulnerability curve to embolism (Petit-Cailleux et al, 2021; Supplementary Appendix 2). To simulate budburst, we used the UniChill model (Chuine et al, 1999) in its version described in Gauzere et al (2017).…”

“…Note that we did not directly simulate mortality due to drought and frost damage with CASTANEA because the thresholds in PLC, NSC, and FD triggering mortality are unknown. Instead, to evaluate the risk of mortality, and compare it between scenarios and species, we computed the relative values of three CASTANEA output variables over all the simulated period, as in Petit-Cailleux et al (2021).…”

Genetic signatures of divergent selection in European beech (Fagus sylvatica L.) are associated with the variation in temperature and precipitation across its distribution range

“…Beech putative glacial refugia and colonization routes have been identified based on very detailed pollen records and genetic population surveys with chloroplast and isozyme markers (Magri et al, 2006). Beech is known to be highly sensitive to summer droughts (Aranda et al, 2015; Knutzen, Dulamsuren, Meier, & Leuschner, 2017), and, to a lesser extent, to late frosts (Kreyling et al, 2014; Petit-Cailleux et al, 2020). Genetic variation has been investigated at various climate-related phenological traits (Gárate-Escamilla, Hampe, Vizcaíno-Palomar, Robson, & Benito Garzón, 2019; Gauzere, Klein, Brendel, Davi, & Oddou-Muratorio, 2020; Gömöry & Paule, 2011; Kramer et al, 2017; Vitasse, Delzon, Bresson, Michalet, & Kremer, 2009), physiological or morphological traits (Bresson, Vitasse, Kremer, & Delzon, 2011; Hajek, Kurjak, von Wühlisch, Delzon, & Schuldt, 2016; Wortemann et al, 2011) and performance traits (Gárate-Escamilla et al, 2019).…”