Tree and ecosystem responses to four years of in situ CO2 enrichment at the Swiss treeline

Following four years of CO2 enrichment (2001-2004) of trees and their understory dwarf
heath, and a one-time tree defoliation treatment in the spring of the second year at the
Swiss treeline FACE site on Stillberg (Davos, Switzerland), this dissertation summarizes
responses from both the above- and below-ground components of this treeline ecosystem.
At the tree physiological level (Handa et al. 2005, chapter 2), we found that
elevated CO2 enhanced photosynthesis in both Larix decidua and Pinus uncinata by ca.
40% and led to increased nonstructural carbohydrate concentrations in the needles of both
species, but to no significant decrease in stomatal conductance. There was no evidence
for photosynthetic downregulation in either of the conifer species. Irrespective of CO2
concentration, defoliation in both species stimulated photosynthesis (Larix, +7 %; Pinus
+52%) and increased stomatal conductance (Larix, +42%; Pinus, +108%) in remaining
current-year needles in the treatment year and reduced leaf nitrogen concentration (-11%
in Larix only) in the year following defoliation. These results are by and large consistent
with what has been observed in multiple other CO2 enrichment experiments to date
(Ceulemans et al. 1999, Norby et al. 1999, Nowak et al. 2004, Zotz et al. 2005) and the
strong physiological effects on the trees from the carbon source removal treatment
highlight how an extreme disturbance can impact the tree’s carbon budget.
Despite the c. 40% stimulation of photosynthesis in response to CO2 enrichment,
this did not translate into carbon that is purely available for growth regardless of whether
one looks at the shoot or stem increment growth records for either of the studied tree
species. In response to elevated CO2, we observed a consistent positive growth response
in Larix evident both in the annual shoot increment record (c. +20-30%; Handa et al.
2005, chapters 2&3) and the stem increment record (+41%; when cumulatively integrated
over four years and measured relative to four years of pre-treatment measurements;
Handa et al. 2006, chapter 3). The increase in radial stem wood growth was the result of
more latewood production, in particular, the formation of larger tracheids, rather than a
greater number of cells. In contrast, both of these lines of evidence (shoot and stem
increment record) showed no positive growth response of Pinus trees, with the exception
of the very first year of shoot increment data (Hättenschwiler et al. 2002, chapter 5). Our studies underline, yet again, how CO2 effects on plants show strong species specificity
(Loehle 1995), and how any meaningful study attempting to address ecosystem
responses, must consider all its key players and account for species diversity (Körner et
al. 2005). Defoliation led to a pronounced decrease in annual ring width of both species,
marked in particular by less latewood production in the treatment as well as subsequent
year, underlining again the importance of how a biotic interaction within the system
might completely modify ecosystem responses in a changing global environment
(Zvereva & Kozlov 2006).
Plants are frequently observed to increase carbon allocation to below-ground
sinks and particularly, to accelerate fine root turnover in response to elevated CO2
concentration. Our study shows that in this natural system, no change in response to
elevated CO2 exposure occurred. There was no difference in total root standing crop after
four years, in new root production measured over three years and also no effect on root
decomposition measured over 26 months (Handa et al. 2008, chapter 4). The lack of
positive growth response below-ground contrasts with the sustained four year aboveground
growth response of Larix decidua, but is in line with the lack of positive aboveground
growth response of the later successional Pinus uncinata trees and that of some of
the understory dwarf shrubs (Zumbrunn 2004). Multiple studies have reported positive
root growth responses to elevated CO2 concentrations, although very few have been
conducted in the field, have exceeded a study duration >1 year or have used mature trees
(Norby & Jackson 2000, Tingey et al. 2000). Root quality measurements indicated that
elevated CO2 significantly increased starch concentration, but there was no change in N
concentration or in dehydrogenase activity. Other studies have also shown higher starch
concentration (Janssens et al. 1998), but also lower N content in roots under elevated CO2
(Janssens et al. 1998, Pregitzer et al. 2000, Wan et al. 2004). However, this result is
certainly not ubiquitous (Tingey et al. 2003, King et al. 2005). Finally, our stable isotope
data indicate that only ca. 30% of the new carbon was incorporated into new roots
indicating a rather slow root turnover in this system.