Abstract

1
Northern Eurasian wetlands and the carbon cycle
Model estimates of carbon storage and methane
emissions Theodore J. Bohn1, KrishnaVeni
Sathulur2, Erika Podest3, Dennis P. Lettenmaier1,
Laura C. Bowling2, Kyle McDonald3 1University of
Washington, Seattle, Washington, USA 2Purdue
University, Lafayette, Indiana, USA 3JPL-NASA,
Pasadena, California, USA American Geophysical
Union Fall Meeting, San Francisco, CA, USA, Dec
10-15 2006
Abstract The Eurasian Arctic drainage
constitutes over ten percent of the global land
area, and stores a substantial fraction of the
terrestrial carbon pool in its soils and boreal
forests. Specifically, boreal forests in this
region constitute an estimated carbon sink of 0.5
Pg/y. However, assessments of carbon storage and
fluxes in this region, and their role in climate
change, vary considerably due to large
uncertainties in the extent of wetlands, which
both store carbon as peat and emit carbon as
methane. Accurate estimates of wetland extent
have been confounded by insufficient resolution
of satellite imagery and poor coverage of in situ
observations. In this study we refine these
estimates of wetland extent, carbon storage, and
methane emissions using a system of linked
large-scale models of hydrology, terrestrial
carbon dynamics, and methane emissions.
Large-scale hydrology comes from the Variable
Infiltration Capacity (VIC) hydrological model,
which includes an updated lake/wetland
parameterization that estimates the water table
depth as a function of both lake level and
wetland soil moisture. Fast ecosystem processes
such as photosynthesis and respiration are
simulated via the Biosphere Energy-Transfer
Hydrology (BETHY) terrestrial carbon model.
Methane emissions in areas of open water or
saturated soil are simulated with the
Walter-Heimann (WHM) methane model. We validate
this modeling system with respect to in situ
observations of soil moisture and temperature,
and fluxes of CO2 and methane at flux towers at
Fyodorovskoe, Russia, over the period 1998-1999.
2. Model Validation Fyodorovskoe Flux Towers
c. CO2 Flux components
Figures 2.c.1 and 2.c.2 show simulated, observed,
and inferred 5-day average carbon fluxes for the
two sites, respectively. Since the flux towers
measure only net CO2 flux from the atmosphere
(net ecosystem exchange, or NEE), we inferred the
actual respiration and NPP by assuming that
night-time NEE is representative of the average
soil respiration rate throughout the day, and
subtracting this from day-time NEE to obtain NPP.
Simulated and results agree with observations in
the general shape of the seasonal cycle. Several
patterns are evident at the two sites first,
observed NEE exhibits considerable scatter during
the growing season, despite having been
aggregated to 5-day averages. Both our inferred
respiration and NPP exhibit this scatter, but
examination of the original half-hourly record
shows that night-time NEE is considerably more
variable than day-time NEE, implying that our
inferring daily respiration from night-time NEE
may be subject to large errors. These
fluctuations may arise from advection via
turbulent fluxes (Alexander Oltchev, pers.
comm.), or alternatively they might occur in
response to precipitation events, in which
infiltrating water forces accumulated CO2 out of
soil pore spaces (Eric Wood, pers. comm.).
Second, BETHY appears to be under-simulating
respiration at the forest site and
over-simulating respiration at the bog site (this
is more clearly expressed in scatter plots 2.c.3
and 2.c.4, for the forest and bog sites,
respectively). This may be a matter of incorrect
vegetation or soil parameters.
2.c.2.
2.c.1.
Located in the Central Forest Biosphere Reserve
in Russias upper Volga basin, the Fyodorovskoe
flux towers have been in operation since 1998.
Meteorological and eddy flux variables have been
recorded at both bog and forest sites. Here we
present the results of point tests in which
observed meteorological forcings over the period
1998-1999 (with a 3-year spin-up) drove both VIC
and a stand-alone version of BETHY. VICs daily
estimates of soil temperature and water table
position, and BETHYs daily estimates of net
primary productivity (NPP), were used as inputs
to the Walter-Heimann methane model (WHM). The
results are shown below.
1. Modeling Approach

• Land Surface Hydrology Model
• Variable Infiltration Capacity (VIC) Model
  (Liang et al. 1994)
• water and energy balance closure
• macroscale
• spatially-distributed
• land cover classification sub-grid variability
• recent additions for cold land processes
  (Cherkauer et al. 2003)
• implemented in arctic regions by Bowling et al.
  (2000) and Bowling et al. (2003)
• lake energy balance component builds on work of
  Hostetler and Bartlien (1990) and Hostetler
  (1991)
• lake/wetland model (Bowling, 2002) handles
  changes in lake extent

2.c.3.
2.c.4.
a. Soil Temperature
2.d.1 Annual Carbon Fluxes
Figures 2.a.1 and 2.a.2 show simulated and
observed soil temperature at 15, 50, and 100 cm
depths, for the forest and bog sites,
respectively. The results agree quite well with
observations at shallower depths, but at deeper
depths VIC appears to have a larger seasonal
cycle than the observed temperatures. This may
result from inaccurate soil parameters we are
still optimizing the calibration here.
2.a.1.
2.a.2.
d. Annual Carbon Fluxes
Annual total fluxes were estimated for each site,
as shown in table 2.d.1. If we neglect the
export of DOC leached from the soil, we can
assess whether the systems are sinks or sources
of atmospheric carbon. Our simulations indicate
that the old forest site is a net sink of both
atmospheric CO2 (459 g C/m2y) and methane (1250
mg C/m2y), while the bog site is a net source for
both CO2 (134 g C/m2y) and methane (433 mg
C/m2y). The CO2 fluxes run counter to our
expectations, but are consistent with BETHYs
under-simulating respiration at the forest site
and over-simulating respiration at the bog site.
However, the methane fluxes, calculated by WHM,
are consistent with our expectation that the
shallow water table during the growing season can
lead to stronger methane emissions.

• Model Framework
• based on framework of Joint Simulation of
  Biosphere Atmosphere Coupling (JSBACH) at Max
  Planck Institute, Hamburg
• represents feedbacks between the physical
  climate system and land surface processes
• modular framework allows components of land
  surface model to be run offline (this project) or
  online
• fast vegetation processes BETHY
• slow vegetation processes LPJ
• combination of land surface, photosynthesis and
  plant respiration schemes (VICBETHY) forms the
  basic coupled model LPJ describes slow changes
  in the distribution of vegetation

Figure 2 Model framework used in this study.
-
-
Conclusions / Future Work

• Future Work
• Continue development of the parameterization of
  spatial variation of the water table in VIC
• Finish the linking of VIC, BETHY, LPJ, and the
  Walter-Heimann methane model
• Add simulations of DOC leaching and aquatic NPP
• Validate these models against historical
  observations
• Validate landcover classifications against in
  situ observations
• Use climate model outputs to drive predictions of
  future lake/wetland extent and carbon cycling in
  Northern Eurasia over the next century

• While this is a work in progress, we can make the
  following conclusions
• Although further refinement is needed, we can
  make reasonable predictions of carbon fluxes in
  forests and bogs.
• Using NEE alone to validate a carbon budget can
  be somewhat imprecise, because it is the
  difference between two terms with large
  variances. Simultaneous measurements of several
  flux terms (e.g NEE, NPP, Rh, DOC export, etc.)
  are essential for constraining errors in carbon
  budgets.

b. Water Table and Methane
Figures 2.b.1 and 2.b.2 show water table position
and methane emissions for the two sites. While
both sites have relatively shallow water tables,
the bog sites water table remains shallow for a
greater portion of the beginning and end of the
growing season, resulting in larger spikes in the
methane emissions curve.
2.b.1.
2.b.2.

• Methane Model
• Walter and Heimann (2000) with modifications
  described in Walter et al (2001a )
• soil methane production, and transport of
  methane by diffusion, ebullition, and through
  plants modeled explicitly
• methane production occurs in the anoxic soil
  bottom of the soil column to the water table
• methane production rate controlled by soil
  temperature and NPP
• time evolution of soil temperature will come
  from VIC

REFERENCES Bowling, L.C., D.P. Lettenmaier and
B.V. Matheussen, 2000. Hydroclimatology of the
Arctic drainage basin, in The freshwater budget
of the Arctic ocean, E.L. Lewis et al. (eds),
Kluwer Academic Publishers, The Netherlands, 57-
90. Bowling, L.C., D.P. Lettenmaier, B.
Nijssen, L.P. Graham, D.B. Clark, M. El Maayar,
R. Essery, S. Goers, Y.M. Gusev, F. Habets, B.
van den Hurk, J. Jin, D. Kahan, D. Lohmann, X.
Ma, S. Mahanama, D. Mocko, O. Nasonova, G.
Niu, P. Samuelsson, A.B. Shmakin, K. Takata, D.
Verseghy, P. Viterbo, Y. Xia, Y. Xue and Z. Yang,
2003. Simulation of high latitude processes in
the Torne-Kalix basin PILPS Phase 2(e) 1
Experiment description and summary
intercomparisons, Global and Planetary Change,
38, 1-30. Cherkauer, K. A., L. C. Bowling and D.
P. Lettenmaier, 2003. Variable Infiltration
Capacity (VIC) Cold Land Process Model Updates,
Global and Planetary Change, 38, 151-159,
2003. De Grandi, G., F. Achard, D. Mollicone,
and Y. Rauste, 2003. The GBFM radar mosaic of the
Eurasian taiga A groundwork for the bio-physical
characterization of an ecosystem with relevance
to global change studies, Proc. IGARSS 2003,
Toulouse, France. Golubev, V.S., N.A.
Speranskaya, and K.V. Tsytsenko, 2003. Total
evaporation within the Volga River Basin and its
variability, Russian Meteorol. Hydrol., 7, 89-99.
Hostetler, S.W. and P.J. Bartlein, 1990.
Simulation of lake evaporation with application
to modeling lake level variations of
Harney-Malheur Lake, Oregon, Water Res. Res., 26,
2603-2612. Hostetler, S.W., 1991. Simulation of
lake ice and its effects on the late-Pleistocene
evaporation rate of Lake Lahontan, Climate
Dynamics 6, 43-48. Liang, X., D. P. Lettenmaier,
E. F. Wood, and S. J. Burges, 1994. A Simple
hydrologically-based model of land surface water
and energy fluxes for GSMs, J. Geophys. Res., 99,
14,415-14,428. Schlosser, C.A., A. Robock, K.Y.
Vinnikov, N.A. Speranskaya, and Y. Xue, 1997.
18-Year land-surface hydrology model simulations
for a midlatitude grassland catchment in Valdai,
Russia. Mon. Weather Rev., 125,
3279-3296. Schlosser, C.A., A.G. Slater, A.
Robock, A. J. Pitman, K. Y. Vinnikov, A.
Henderson-Sellers, N. A. Speranskaya, K.
Mitchell, and the PILPS 2(d) contributors, 2000
Simulations of a boreal grassland hydrology at
Valdai, Russia PILPS Phase 2(d). Mon.
Weather Rev, 128, 301-321. Vinnikov, K.Y., A.
Robock, N.A. Speranskaya, and C.A. Schlosser,
1996. Scales of temporal and spatial variability
of midlatitude soil moisture. J. Geophys. Res.,
101, 7163-7174. Walter, B. P., and M. Heimann,
2000. A process-based, climate-sensitive model
to derive methane emissions from natural
wetlands Application to five wetland sites,
sensitivity to model parameters, and climate,
Global Biogeocheical. Cycles 14,
745765. Walter, B.P., M. Heimann, and E.
Matthews, 2001a Modeling modern methane
emissions from natural wetlands 1. Model
description and results, J. Geophys. Res. 106,
34,189 34,206 Walter, B.P., M. Heimann, and E.
Matthews, 2001b Modeling modern methane
emissions from natural wetlands 2. Interannual
variations 1982-93, J. Geophys. Res. 106, 34,207
34,219.
Depth (cm)
Depth (cm)