| Popis: |
Stable isotopes can diagnose the response of plants to changing climate as the performance of trees in past climatic conditions is archived in the stable carbon and oxygen isotope composition (δ13C and δ18O, respectively) of tree rings. To take advantage of these records, understanding the formation of isotopic signals in newly assimilated photosynthates is necessary. Despite a voluminous literature, there exists a gap between the model- and data-oriented studies, which if welded together would benefit this line of inquiry. A unique dataset covering two growing seasons in a boreal Scots pine stand situated in Southern Finland (61.9°N, 24.3°E) is employed and is accompanied with mechanistic modeling driven by environmental conditions. Data includes: (i) shoot gas exchange of vapor, CO2 and its δ13C composition, (ii) δ13C in needle bulk sugar and sucrose alone, (iii) δ18O in water in precipitation, soil, twigs and needles, and (iv) δ18O in needle bulk sugar. Overall, observed exchange rates and isotopic composition of fluxes as well as in water and sugar pools were well reproduced using the model. We further address challenges common to the analysis of isotopic signals. Firstly, time scales and integration over them is an unavoidable challenge of data sampled at different intervals, representing either snapshots or a longer history of processes. As an example of this, we illustrate that δ18O in needle water reacts instantaneously to environmental conditions, while the δ18O signal in needle sugars is an integration over time, and thus relating the latter to instantaneous environmental conditions is less evident. Given that tree-ring studies are more and more focused on intra-annual variation in δ13C and δ18O, integration over time scales cannot be neglected. Second, using model sensitivity analysis, we showcase the relative importance of environmental drivers on the variation in δ13C and δ18O – the typical aim of empirical research and paleoclimatological reconstruction. It is commonly acknowledged that the main environmental driver of δ13C or δ18O variation can differ between sites and time periods. At the study site here, the variation in δ18O seems solely driven by relative humidity, but we can, for instance, show that this would change if the δ18O signal of source water varied considerably. We are of the opinion that illustrating such points with a model-data fusion approach is a necessary (but not sufficient) first step to bridge the gap between modeling and empirical approaches, and fostering further interpretation of isotopic signals in trees. |
