Drought effects on carbon and nutrient dynamics in Scots pine : homeostasis, thresholds and tipping points

Species of the tree genus Pine (Pinus L.) exist all over the world and no other group contains so many attractive forms (Curtis & Bausor, 1943) . Scots pine (Pinus sylvestris L.) is currently the most widely distributed pine and occurs throughout all of Eurasia. In the central alpine valleys, Scots pine is growing at the dry border of its distribution range, which involves overcoming periods with extreme low water availability. Although the species is known for its ability to grow on dry and nutrient poor soils, several extreme droughts during the last two decades have caused a 50% dieback of Scots pine in the dry valleys of the Central Alps in Switzerland. The ability of trees to survive drought is determined by their initial health and their resilience to drought, as well as on the characteristics of a drought event – i.e. timing, duration and intensity. The mechanisms underlying drought-induced mortality are still unclear, as well as the recovery process after soil rewetting. Furthermore, possible mitigation or aggravation of drought effects by elevated nutrient availability in the soil has not been studied before. The carbon (C) balance in trees is used as an indicator for C assimilation, growth, defense and storage processes. When trees are exposed to drought, to changes in soil nutrition or sudden defoliation, the C balance may change. In this thesis, the main objective was thus to combine effects of drought and fertilization to study the C and nitrogen (N) dynamics in Scots pine trees.
In the first chapter, I give an overview of the state-of-the-art in research on drought-affected C and N dynamics in trees. The aim of the second chapter was to assess the effects of long-term drought release on growth and non-structural carbohydrate (NSC) concentrations of adult P. sylvestris trees. A long-term (13 years) irrigation experiment was conducted in the Pfynwald, a Scots pine dominated forest located at the dry distribution margin of the species in southern Switzerland. I measured growth, NSC, N and phosphorus (P) concentrations, as well as the natural abundance of 13C isotopes on trees with different leaf area in control and irrigation plots. Irrigation resulted in higher growth rates and carbon isotope discrimination, but did not alter NSC levels. Growth and NSC decreased with lower leaf area in both control and irrigated trees, but NSC did not correlate with leaf-level gas exchange indices such as foliar δ13C, which is an indicator for water use efficiency, N or P, which are both stimulants of photosynthesis. Trees with initially low leaf area had limited ability to respond to the long-term irrigation, indicating a legacy effect of previously low crown condition. The NSC constancy across treatments suggests that carbohydrate storage may stay constant when changes in climate are slow enough to allow acclimation. Moreover, total leaf area, rather than leaf gas exchange per unit leaf area, drives variation in whole-tree carbohydrate dynamics in this system.
The main focus of the third chapter was the mitigation or aggravation of drought effects by nutrient availability in the soil. Three year-old P. sylvestris saplings were exposed to drought during two subsequent years, using four different water and two soil nutrient regimes, and drought was released thereafter. In addition, partial and full needle removal was performed in order to assess effects of changes in source:sink ratio. Biomass, leaf gas exchange and tissue NSC were measured during and after the first and second growing season. Extreme drought reduced stomatal conductance, photosynthesis, biomass and NSC, whereas intermediate drought only slightly affected biomass and NSC. Defoliation stimulated photosynthesis and fertilization increased growth and root biomass fraction, but mainly in the two intermediate drought levels. Only extreme drought pushed P. sylvestris trees to mortality. The third chapter concludes that tree mortality under severe drought periods will not be mitigated, but that the effects of low intensity drought stress could be compensated by increased nutrient availability and decreased source:sink ratio.
The aim of the fourth chapter was to assess the C and N allocation underlying the biomass changes that were found in chapter 3. I hypothesized that, during drought, increased soil nutrient availability stimulates root metabolism and carbon allocation to belowground tissues under drought stress. I therefore conducted a 15N and 13C labelling experiment in July and August 2016 respectively, on the saplings described above. 15N labelling was conducted with fertilized saplings from all water regimes, while 13C labelling was only conducted with saplings (both nutrient regimes) from two out of four water regimes (well-watered and mild drought). I assessed the abundance of 15N and 13C in the roots, stem and needles after the first growing season and during the second year. C uptake was slightly lower in drought stressed trees, and extreme drought inhibited largely the N uptake and transport. Carbon allocation to belowground tissues was decreased under drought, but not in combination with fertilization. The results indicate a potential positive feedback loop, where fertilization improved the metabolism and functioning of the roots, stimulating source activity and hence C allocation to belowground tissues. We can thus conclude that soil nutrients might play an important role in mitigating drought stress of trees.
Overall this thesis shows that the impairment of tree functioning and mortality can be explained with thresholds: long-term drought causes a reduction in tree vigor and leaf area, and if a threshold of approximately 60 – 70% loss of leaf area is reached, trees may follow a trajectory towards mortality, even if drought is released in the soil. In the controlled experiment, soil moisture thresholds were visualized. The impairment of C allocation belowground under mild drought, the reduction of NSC in and impairment of 15N uptake by the roots under extreme drought indicate that roots might be the first tissue to lose function and eventually die off during drought stress. Additional nutrient supply can sustain root functioning under drought, indicating that soil moisture tipping points are not fixed, but can be modified. In general, trees have a strongly coordinated supply – demand regulation for C and N, enabling homeostatic C balances as long as changes in climate are slow or mild enough for trees to acclimate.