Northern peatlands are a major terrestrial carbon (C) store, with an annual sink of
0.1 Pg C yr-1 and a total storage estimate of 547 Pg C. Northern peatlands
are also major contributors of atmospheric methane, a potent greenhouse gas. The
microtopography of peatlands helps modulate peatland carbon fluxes; however, there is
a lack of quantitative characterizations of microtopography in the literature. The lack
of formalized schemes to characterize microtopography makes comparisons between studies
difficult. Further, many land surface models do not accurately simulate peatland C
emissions, in part because they do not adequately represent peatland microtopography
and hydrology. The C balance of peatlands is determined by differences in C influxes
and effluxes, with the largest being net primary production and heterotrophic respiration,
respectively. Tree net primary production at a treed bog in northern Minnesota represented
about 13% of C inputs to the peatland, and marks tree aboveground net primary production
(ANPP) as an important pathway for C to enter peatlands. Tree species Picea mariana
(Black spruce) and Larix Laricina (Tamarack) are typically found in wooded
peatlands in North America, and are widely distributed in the North American boreal zone.
Therefore, understanding how these species will respond to environmental change is needed
to make predictions of peatland C budgets in the future. As the climate warms, peatlands
are expected to increase C release to the atmosphere, resulting in a positive feedback
loop. Further, climate warming is expected to occur faster in northern latitudes compared
to the rest of the globe. The Spruce and Peatland Responses Under Changing Environments
(SPRUCE; https://mnspruce.ornl.gov/) manipulates temperature and CO2 concentrations
to evaluate the in-situ response of a peatland to environmental change and is
located in Minnesota, USA. In this dissertation, I documented surface roughness metrics
for peatland microtopography in SPRUCE plots and developed three explicit methods for
classifying frequently used microtopographic classes (microforms) for different scientific
applications. Subsequently I used one of these characterizations to perform a sensitivity
analysis and improve the parameterization of microtopography in a land surface model that
was calibrated at the SPRUCE site. The modeled outputs of C from the analyses ranged from
0.8-34.8% when microtopographical parameters were allowed to vary within observed ranges.
Further, C related outputs when using our data-driven parameterization differed from
outputs when using the default parameterization by -7.9 - 12.2%. Finally, I utilized TLS
point clouds to assess the effect elevated temperature and CO2 concentrations
had on P. mariana and L. laricina after the first four years of SPRUCE
treatments. I observed that P. mariana growth (aboveground net primary production)
had a negative response to temperature initially, but the relationship became less pronounced
through time. Conversely, L. laricina had no growth response to temperature initially,
but developed a positive relationship through time. The divergent growth responses of P. mariana
and L. laricina resulted in no detectable change in aboveground net primary production at the
community level. Results from this dissertation help improve how peatland microtopography is
represented, and improves understanding of how peatland tree growth will respond to environmental
change in the future.
The essential carbon service provided by northern peatlands
