High Spatial and Temporal Resolution Ecosystem CO 2,

High Spatial and Temporal Resolution Ecosystem CO 2,

High Spatial and Temporal Resolution Ecosystem CO 2, Latent Heat, and Sensible
Heat Fluxes in the SGP: Tools for Cloud Models, CLASIC, and NACP
William J. Riley1 ([email protected]),Sbastien C. Biraud1, Marc L. Fischer1, Margaret S. Torn1, Joseph A. Berry2
1: Earth Sciences Division, Lawrence Berkeley National Laboratory, Berkeley, CA
2: Dept. Global Ecology, Carnegie Institution of Washington, Stanford, CA

RESULTS

INTRODUCTION

Comparing climate interpolation with ECOR measurements:
Estimating spatially distributed and temporally resolved
ecosystem fluxes is important for ARM cloud modeling, the 2007
CLASIC experiment, and for the North American Carbon
Program. Because the land surface is very spatially
heterogeneous in the ACRF (LAI and vegetation cover change on
scales of about 200 m), characterizing vegetation type and
characteristics is critical to accurately estimating CO 2 and energy
surface exchanges with the atmosphere. We describe here a
methodology to estimate surface energy fluxes and net ecosystem
CO2 Exchange (NEE) continuously over the Southern Great
Plains, using (1) meteorological forcing data from Mesonet
facilities; (2) the USGS soil database; (3) NEXRAD 4 km
precipitation estimates; (4) eddy covariance data from several
sites over several years in the ACRF; and (6) a thoroughly tested
land-surface model (ISOLSM). This approach allows us to
estimate fluxes in periods and areas where meteorological forcing
data are unavailable.
In this poster we present our method, compare site level
model estimates of CO2 and energy exchanges, compare regional
model predictions to measurements made with eddy correlation
flux systems, and discuss estimates of regional means and
variability.

MATERIALS & METHODS
We incorporate meteorological data from distributed
Oklahoma and Kansas Mesonet sites (Doran et al., 1998),
satellite derived vegetation cover and characteristics, and
NEXRAD precipitation estimates into a distributed Land Surface
Model (ISOLSM, Riley et al. (2002, 2003)) of fluxes between
ecosystems and the atmosphere. The meteorological datasets are
compiled by ARM and the MODIS satellite data are available
from NASA.
Interpolation of Climate Measurements:
We apply the approach of Doran et al. (1998) to generate
the model climate forcing (wind speed, downward solar radiation,
and air temperature, pressure, and humidity) from the Mesonet
data. The interpolation procedure provides the climate forcing for
ISOLSM at a user-specified resolution. Given the spacing of the
Mesonet stations, the minimum justifiable resolution is about 10
km, which we use in the results shown here. For example, Figure
1 shows the result of the interpolation for air temperature. Note
that the density of stations in KS is much lower than in the OK
portion of the ACRF. We calculated precipitation inputs from the
4 km NEXRAD production (Figure 2).

Figure 3 shows interpolated (from the Mesonet data) and
measured relative humidity, air temperature and wind speed at
one wheat field in April 2003. The interpolated values are in
good agreement with independent measurements (i.e. from
locations not used in the interpolation). Similarly good
agreement was obtained for our other eddy flux sites.

Figure 2: Precipitation varies strongly from South to North, with about double
the precipitation falling in the Southeast as compared to the Northwest.

Figure 6: Regional mean (blue) NEE across the ARM-SGP domain in June
2003. Red line corresponds to plus or minus one standard deviation around
mean NEE.

ISOLSM Land Surface Model:
ISOLSM (developed from LSM1.0, Bonan, 1996, 1997) is a
big-leaf land surface model that simulates CO2, H2O, energy
and isotope (e.g., C18OO, HDO, H218O) fluxes between ecosystems
and the atmosphere. ISOLSM simulates aboveground fluxes of
radiation, momentum, sensible heat, and latent heat; energy and
water fluxes below ground; and coupled CO2, H2O, and isotope
exchange between soil, plants, and atmosphere. Soil hydraulic
characteristics are determined from sand, silt, and clay content.
We modified the surface types in ISOLSM to correspond to the
dominant land cover in the ARM-SGP domain, and calibrated the
model to these land covers using our eddy covariance
measurements over several years. ISOLSM has been applied in a
number of studies in the ACRF (Riley et al., (2002, 2003); Cooley
et al. (2005); Riley (2005); Lai et al. (2006); Still et al. (2005,
2007)) and in other sites and regions (Henderson-Sellers et al.
(2006); Aranibar et al. (2006), Riley et al. (2005)).
We used 250 m MODIS NDVI data to estimate LAI using
the approach of Sellers et al. (1996). Vegetation cover type was
estimated using archetypal LAI profiles for each vegetation type
and the MODIS data. Average percent vegetation cover for four
quadrants (NW, NE, SW, and SE) in the ACRF determined from
the MODIS NDVI compared well with the 2002 Census data
(Figure 3), and captured the large spatial variability in C3 winter
crop versus pasture prevalence. To ensure that the simulations are
tractable computationally, we sub-sample ten of each vegetation
type cells per user-defined macrocell (here 10 km).

For June, 2003, regional mean NEE varied between -15 and +5
mol m-2 s-1 diurnally (Figure 6). Variations in midday NEE result
from differences in meteorological forcing, consequent soil
moisture, and vegetation characteristics. Spatial heterogeneity in
NEE was largest midday (up to 20 mol m-2 s-1 ) and smaller at
night (~2 mol m-2 s-1 ). In June, a majority of the cumulative
carbon uptake was by C4 crops and pasture. The other vegetation
types did not contribute substantially to cumulative uptake.
During April, the majority of the uptake over the region was by
winter C3 crops (i.e., winter wheat).

Figure 4: Comparison of interpolated (red line) and measured (blue dots):
(a) net radiation; (b) sensible heat; and (c) latent heat.

Simulated and measured NEE:

CONCLUSIONS

The surface energy balance (latent heat, sensible heat, long-wave
radiation) varied greatly over the domain, depending on the type
of vegetation and its phenology.
Characterizing vegetation type using archetypal LAI profiles and
MODIS data resulted in good predictions as compared to census
county level data.
Comparisons were good between interpolated and independently
measured climate forcing
Comparisons were good between predicted and measured CO2,
latent heat, and sensible heat fluxes at our portable eddy
covariance sites.
The forcing fields for the modeling work will be made available
via the ARM website.
Figure 5: Time series of measured and simulated NEE of wheat (April 2003),
pasture, and sorghum (June 2003). Red lines and blue dots correspond to
simulated and measured CO2 fluxes, respectively.

Figure 3. Predicted and USDA Census data for vegetation cover for the
four quadrants surrounding the CF

Eddy flux measurements:

Figure 1: Interpolated air temperature field over the ARM-SGP domain on
April 3, 2003 at 9:00 UT. Open circles show the location of Mesonet
platforms and filled square shows the ARM Central Facility.

Comparable large variation across the ACRF were also seen in
latent and sensible heat fluxes, implying that estimating regional
exchanges requires accurate characterization of spatial
heterogeneity in vegetation type, vegetation characteristics, and
meteorological forcing.

We compared simulated and ongoing measurements of land
surface-atmosphere exchanges of CO2 and energy as well as
ancillary data from pasture (C3/C4 grass mix), wheat (winter C3
crop), and sorghum (C4 crop) fields, which are the dominant crop
types in the ARM-SGP domain (Fischer et al., 2007). These
measurements are made with portable systems comprised of a
sonic anemometer, an open-path infrared gas analyzer (IRGA),
and a set of meteorological instruments that monitor net and
photo-synthetically active radiation, air temperature, relative
humidity, precipitation, soil heat flux, and soil moisture and
temperature. The measurement height is 4 m above the ground,
allowing a minimum of ~3 m between the top of the canopy and
the IRGA and sonic anemometer for crops included in this study.
(see http://www.arm.gov/instruments/carbon.stm for measurement
description).

Figure 5 shows a comparison between simulated and measured
net ecosystem CO2 flux for 3 crop types (wheat, pasture, and
sorghum). We calibrated the model for each vegetation type
using a 2 approach with the maximum carboxylation rate (Vmax)
and soil organic matter content.
Midday NEE in April wheat was underestimated for the first
two weeks. Estimates during the remainder of the month were
good. NEE predictions in June for the pasture were accurate;
peak uptake was approximately uniform across the month.
Estimates in June for the sorghum field underestimated NEE for
the first two weeks, and then accurately captured the decrease in
NEE as the plants senesced. Overall, NEE during the nighttime
was more accurately predicted than during the daytime. This
result occurs because of the relative simplicity of mechanisms
operating at night versus day.

REFERENCES
Aranibar, J.N., J.A. Berry, W.J. Riley, D.R. Bowling, J.R. Ehleringer, D.E. Pataki, and B.E. Law (2005) Modeling environmental
controls of carbon isotope discrimination, carbon, and energy fluxes at the canopy scale in a semi-arid pine forest, 12, 710730, doi:
10.1111/j.1365-2486.2006.01121.x, Global Change Biology.
Billesbach, D. P., M. L. Fischer, M. S. Torn and J. A. Berry. A Portable Eddy Covariance System for the Measurement of
Ecosystem-Atmosphere Exchange of CO2, Water Vapor, and Energy. Journal of Atmospheric and Oceanic Technology 21(4): 639650, 2004.
Bonan, G. B., A land surface model (LSM version 1.0) for ecological, hydrological, and atmospheric studies: Technical description
and user's guide, pp. 150, NCAR, Boulder, CO, 1996.
Bonan, G. B., K. J. Davis, D. Baldocchi, D. Fitzgerald, and H. Neumann, Comparison of the NCAR LSM I land surface model with
BOREAS aspen and jack pine tower fluxes, Journal of Geophysical Research, 102 (C12), 29,065-29,076, 1997.
Christensen, L., W.J. Riley, and I. Ortiz-Monasterio (2006) Nitrogen cycling in an irrigated wheat system in Sonora, Mexico:
Measurements and modeling, DOI 10.1007/s10705-006-9025-y, Nutrient Cycling in Agroecosystems.
Cooley, H.S., W.J. Riley, M.S. Torn, and Y. He, (2005) Effect of harvest on regional climate and soil moisture and temperature, 110,
D03113, doi:10.1029/2004JD005160, JGR Atmospheres.
Doran, J. C., J. M. Hubbe, J. C. Liljegren, W. J. Shaw, G. J. Collatz, D. R. Cook, R. L. Hart. A technique for determining the spatial
and temporal distributions of surface fluxes of heat and moisture over the Southern Great Plains Cloud and Radiation Testbed.
Journal of Geophysical Research-Atmospheres 103(D6): 6109-6121, 1998.
Henderson-Sellers, A., M. Fischer, K. McGuffie, W.J. Riley, G. Schmidt, K. Sturm, K. Yoshimura (2006) Stable Water Isotope
Simulation by Current Land-surface Schemes: Results of iPILPS Phase 1, 51, 34-58, doi:10.1016/j.gloplacha.2006.01.003, Global
and Planetary Change.
Lai, C., W.J. Riley, C. Owensby, J. Ham, A. Schauer, and J. R. Ehleringer (2006) Seasonal and interannual variations of carbon and
oxygen isotopes of respired CO2 in a tallgrass prairie: Measurements and modeling results from 3 years with contrasting water
availability, 111, D08S06, doi:10.1029/2005JD006436, J. Geophys. Res.
Riley, W.J., C.J. Still, M.S. Torn, and J.A. Berry (2002) A mechanistic model of H218O and C18OO fluxes between ecosystems and
the atmosphere: Model description and sensitivity analyses, Global Biogeochemical Cycles, 16, 1095-1109.
Riley, W.J., C.J. Still, B.R. Helliker, M. Ribas-Carbo, and J.A. Berry (2003) Measured and modeled 18O in CO2 and H2O above a
tallgrass prairie, 9, 1567-1581, Global Change Biology
Riley, W.J., J.T. Randerson, P.N. Foster, and T.J. Lueker (2005) The influence of terrestrial ecosystems and topography on coastal
CO2 measurements: A case study at Trinidad Head, California, JGR-Biogeosciences, No. G1, G01005, 10.1029/2004JG000007

Regional Mean Ecosystem Exchanges

Riley, W.J. (2005) A Modeling Study of the Impact of the 18O Value of Near-Surface Soil Water on the 18O Value of the SoilSurface CO2 Flux, 69(8), 19391946, Geochimica et Cosmochimica Acta.

Because we sub-sample the 250 m pixels across the
ACRF, regional estimates are made by combining (as mean and
SD) estimates for each vegetation type, and then scaling to the
10 km grid using the relative prevalence of each vegetation type.
Figure 5 shows mean and SD NEE predictions for a three-week
period in early summer.

Still, C.J., W.J. Riley, B.A. Helliker, and J.A. Berry (2005) Simulation of ecosystem oxygen18 CO2 isotope fluxes in a tallgrass
prairie: Biological and physical controls, In Stable Isotopes and Biosphere-Atmosphere Interactions (Eds. Flanagan, L.B., Ehleringer,
J.R. & D.E. Pataki). Elsevier-Academic Press, Physiological Ecology Series.

ACKNOWLEDGEMENTS
This work was supported by the ARM Program, U.S. Department
of Energy Office of Science, Biological and Environmental
Research Program, under Agreement DE-FC03-90ER61010.

Recently Viewed Presentations

  • Workforce Solutions

    Workforce Solutions

    WORKFORCE SOLUTIONS. Westheimer Greeter Dennita Allen. As a flower in the gardenBending toward the sun,Unfolds its tiny petalsOne, by one, by one…So faith expands its beautyUntil at last it growsInto life's lasting flower…The heart's fair perfect rose. Helping with Young...
  • Required rate of return  When valuing assets and

    Required rate of return When valuing assets and

    Firm-specific risk . Also called idiosyncratic risk, asset-specific risk, diversifiable risk. Risk that affects one or a few investments. Can be diversified away so that investors are not rewarded for bearing this risk. Market risk .
  • Chapter 3 Nuclear Radiation - HCC Learning Web

    Chapter 3 Nuclear Radiation - HCC Learning Web

    49Mn 49Cr + 0e 25 24 +1 Gamma Radiation In gamma radiation energy is emitted from an unstable nucleus, indicated by m following the mass number. the mass number and the atomic number of the new nucleus are the same.
  • A phase trajectory is shown for populations of rabbits ( R ...

    A phase trajectory is shown for populations of rabbits ( R ...

    In the absence of wolves, the rabbit population is always 5000 Zero populations Both populations are stable * Title: A phase trajectory is shown for populations of rabbits ( R ) and foxes ( F ). Describe how each population...
  • Principles of Education and Training - Stephen F. Austin ...

    Principles of Education and Training - Stephen F. Austin ...

    Copyright © Texas Education Agency, 2014. The materials found on this website are copyrighted © and trademarked ™ as the property of the Texas Education Agency ...
  • Living, Non-Living, Dead

    Living, Non-Living, Dead

    Non-Living. In order to consider something non-living, it must not have any of the 6 characteristics of living things. Examples: chair, pen, desk, water. or. It was living or dead and has been . processed (chemically altered) Examples: paper, leather...
  • Amena Saeed (H00227813) Mariam Saeed (H00247930) Khawla Abdulla

    Amena Saeed (H00227813) Mariam Saeed (H00247930) Khawla Abdulla

    The cinema in Egypt. Famous Egypt actors are :-Ahmed Helmy-Ahmed El Sakka-Laila Mourad-Faridshawky-Nour El-Sherif-Mohamed Hinidi. Historians have differed in defining the beginning of cinema in Egypt, there are those who say that beginning in 1896 with the premiere film in...
  • Téma: Psychologie inteligence

    Téma: Psychologie inteligence

    Pedagogická psychologie