Tree growth and drought impact the dynamics of C allocation and change the coupling between photosynthesis and respiration in stem and soil

<p>In-depth knowledge about carbon (C) flows within trees and from the trees to forest ecosystem via respiration is essential for accurate modeling of tree growth and C balance. However, significant gaps still exist in our understanding about how trees allocate C for growth and respiration of different tree organs, which makes it difficult to predict the response of forest growth to climate change. A powerful tool to study C allocation within trees is stable C isotope ratio (the ratio of <sup>13</sup>C to <sup>12</sup>C relative to a reference, noted as δ<sup>13</sup>C), as this signal is passed from C sources to C sinks with isotopic fractionation along the pathway. In this study, we monitored the δ<sup>13</sup>C signal of CO<sub>2</sub> fluxes of shoot (A<sub>canopy</sub>), stem (R<sub>stem</sub>) and soil (R<sub>soil</sub>) in a Scots pine (Pinus sylvestris L.) dominated boreal forest in southern Finland for summer 2018, which included a month-long dry period. We also traced the growth of current-year shoots, needles, stem, and fine roots (fibrous and pioneer roots) and the concentrations and δ<sup>13</sup>C of putative substrates (sugars and starch) in phloem and roots of Scots pine over the growing season. We calculated the correlations between substrate concentrations and respiration fluxes, as well as the correlations between δ<sup>13</sup>C of A<sub>canopy</sub> and δ<sup>13</sup>C of R<sub>soil</sub> or δ<sup>13</sup>C of R<sub>stem</sub> with varying time lags from 3 d to 14 d for different tree organ growth periods and the dry period. We found tight couplings between photosynthesis and respiration, when newly assimilated sugars were allocated to stem or roots for growth or for drought response. These couplings include: 1) a synchrony between fibrous root growth and the concentrations of bulk sugars and starch in roots, associated with increases in R<sub>soil</sub> under high root substrate concentrations; 2) promoted nighttime R<sub>stem</sub> under high substrate supply to stem, which is seen as increased phloem glucose to sucrose ratio; 3) shorter time lags between δ<sup>13</sup>C of A<sub>canopy</sub> and δ<sup>13</sup>C of R<sub>stem</sub> under higher stem growth demands; 4) shorter time lags between δ<sup>13</sup>C of A<sub>canopy</sub> and δ<sup>13</sup>C of R<sub>soil</sub> under drought stress than with no water stress. The time lags between δ<sup>13</sup>C of A<sub>canopy</sub> and δ<sup>13</sup>C of R<sub>soil</sub> or δ<sup>13</sup>C of R<sub>stem</sub> being not uniform further implies that tree C allocation patterns are dynamic over the growing season. In addition, the C allocation to stem and roots occurred after full expansion of current-year shoots or needles, reflecting a whole tree C allocation strategy for growth demands of different tree organs, which prioritizes the demands of source organs. We suggest that the dynamics of C allocation in response to tree organ growth and drought stress should be considered in whole tree C allocation models for projecting forest growth under climate change.</p>