Earth Surface Dynamics

Exchanges of Greenhouse Gases, Water Vapor,
and Heat at the Earth's Surface

Atmospheric turbulence transports greenhouse gases (notably: carbon dioxide (CO2), methane, and nitrous oxide), water vapor, and heat between Earth's land and water surfaces and the overlying atmosphere. These exchanges influence climate, viability of ecosystems, distribution of biomes, and the quantity of both surface- and ground-water. Using the eddy covariance method (explained later, below) derived from fundamental micrometeorological theory, the exchanges of greenhouse gases, water vapor, and heat may be directly measured to better understand interactions between Earth's surface and atmosphere.

Lake-atmosphere CO2 exchange


Wind, temperature, humidity, and CO2 concentration sensing instruments on tower in Williams Lake. Raft contains data logging and logistical support equipment.

Researcher adjusting micrometeorological instruments on tower in Williams Lake.

Although lakes occupy only a small fraction of Earth's surface area, study of the CO2 exchange between them and the atmosphere is needed to better define the carbon cycle and to improve our understanding of air-water gas transfer over much larger water bodies such as oceans. Williams Lake (part of the USGS's Interdisciplinary Research Initiative in the Mississippi River headwaters area of north-central Minnesota) was selected for a 3 year (1992-4) study of air-water CO2 exchange. Detailed measurements of lake-water CO2 concentrations were made over the course of the experiment. Eddy covariance instruments were deployed from a tower that was inserted into lake sediments and guyed underwater.

A raft tethered to the tower supplied data logging and logistical support for the equipment. Eddy covariance measurements of CO2 flux and measured change in lake CO2 storage were compared with state of the art models of flux over water. Agreement was inconsistent between the two measurements and the models, bringing into question whether some other mechanisms, not represented in the models, were at work. The models tested are often used to estimate global air-sea CO2 exchange. (See Anderson et al. 1999; Limnology and Oceanograph vol. 44 for further details.)

Forests: Sequestering atmospheric CO2 and humidifying the atmosphere

Boreal forests


Researcher installing micrometeorological instruments on a communications-type tower in a Canadian coniferous forest.
For decades, a large number of scientists have found that after accounting for all known global carbon sinks and sources, a large unknown sink of atmospheric CO2 seems to exist somewhere in the biosphere. BOREAS (BOReal Ecosystem Atmosphere Study) was developed and executed (1993-5) in north-central Canada to determine to what extent boreal forests may be contributing to this missing sink. A team of USGS scientists operated eddy covariance instruments deployed from a tower above a jack pine forest during BOREAS, which measured the net exchange of CO2 and heat, and evapotranspiration (the combination of evaporated and transpired water). USGS researchers joined others from U.S. (NASA, NOAA, EPA) and Canadian (Forestry, Agriculture) governmental agencies and a variety of universities from several countries, employing a variety of measurement methodologies over the mosaic of boreal landscapes, to determine how boreal forests affect climate (through modification of atmospheric greenhouse gas concentrations, heating, humidification, and radiant energy exchange), and vice versa.

In addition to establishing that jack pine forests are a weak, perhaps ephemeral, sink for atmospheric CO2, the USGS team found age dependent differences of carbon sequestration with younger forests drawing about 20% more CO2 than mature stands. The USGS and other BOREAS science teams found that evapotranspiration from these forests was considerably less than originally thought while heat flux to the atmosphere was considerably more. This finding has bearing on the parameterization of these forests in numerical weather prediction models (see J. Geophys. Res. (1997), BOREAS Special Issue, Vol. 102).

U.S. Subalpine forest


Subalpine forest site west of Boulder, Colorado. The Rocky Mountain range in the background is part of the Continental Divide.
Long term (years), continuous, measurements of forest ecosystem-atmosphere exchange of CO2, evapotranspiration, and heat flux will be conducted in a subalpine forest (3048 meters (10,000 feet) MSL), near the Continental Divide, west of Boulder, CO. Though of much smaller scale, this study has similar science goals as were stated above for BOREAS. An additional goal is to improve our fundamental understanding of the character of turbulent transport of heat and gas in mountainous (complex) terrain. Presently, very little is known of forest-atmosphere exchanges of greenhouse gases, heat, and evapotranspiration from montane and subalpine forests.

Infrastructure is currently being emplaced near the Niwot Ridge LTER (Long term Ecological Research) site. This site operates as a part of Ameriflux -a group of individual investigators making eddy covariance measurements in North and South America (see Science (1998) Vol 281. p.506, for more information). Detailed eddy covariance flux measurements, and wind and CO2 concentration profile measurements will be conducted from two 33 m (110') tall towers, individually operated by the University of Colorado and the USGS, in collaboration with NCAR (National Center for Atmospheric Research) researchers and with the cooperation of the U.S. Forest Service. This will be the first Ameriflux site capable of directly measuring nocturnal drainage flow of CO2 flux originating from ecosystem respiration. CO2 flux carried by this flow can be a substantial fraction of the total diurnal and annual forest-atmosphere flux.

CO2 Emissions from Volcanoes

Volcanic CO2 emissions from low-level eruptive activity can pose a threat to life in its vicinity and is a topic of interest in a range of sciences including global climate change and chemical geology. In several areas on Mammoth Mountain, a dacitic volcano on the eastern Sierras of California, soil CO2 concentrations are in excess of 70% or 700,000 ppm (global average atmospheric concentration is about 355 ppm). Very high rates of tree mortality accompany these areas since roots require oxygen for respiration, not CO2.


Horseshoe Lake appears in the foreground at the foot of Mammoth Mountain. The area of high CO2 flux and tree mortality appears at the far shore of the lake.

Trees killed on Mammoth Mountain due to very high concentrations of CO2 in soils.

In a unique application of the eddy covariance method, diffuse gas emissions from the volcano's flank were continuously measured in three pilot studies (1996,'97,'98). These week-long studies afforded the opportunity to evaluate the feasibility of this methodology in the measurement of a strong, heterogeneous, CO2 source in complex terrain. These were the first direct measurements of CO2 flux of volcanic origin. Results indicate that the basic assumptions underlying the eddy covariance technique were preserved in this experiment and that long term, continuous measurements may be made at the site. (Procedures and results are to appear later this year in Chemical Geology.)

Why Was the Eddy Covariance Method Chosen in the Above Studies?

A variety of micrometeorological methods and surface based chambers are available to make gas and heat flux measurements (see Anderson, D.E. 1999. Chapter 9 in Unsaturated Zone Hydrology for Scientists and Engineers (Tindall and Kunkel), Prentice Hall). Considering all methods (above), eddy covariance is the only method in which the exchanges or fluxes of gas and heat between the Earth's surface and the atmosphere may be directly measured for large areas (hectares to 100's of hectares).


Eddy covariance equipment used in some studies mentioned on this page. Left to right in foreground: CO2 and Water Vapor sensor; Sonic anemometer capable of measuring wind velocity and temperature (center); and, Sonic anemometer (chrome color body) capable of sensing only vertical wind and temperature.
A mean flux of gas or heat may be measured with the eddy covariance method when turbulence, generated by the movement of air immediately above the Earth's surface, transports the gas or heat energy across a plane normal to the surface generating it.

Gas or heat flux is determined by calculating a covariance statistic of the fluctuations in vertical wind velocity and the entity of interest (gas concentration or temperature) from a series of sequential measurements made over time (about 30 minutes, typically). To sufficiently capture significant turbulent motions, eddy covariance instrumentation must be capable of accurately sampling wind or gas concentration at least several times per second. The instrumentation may be costly (depending upon needs) and considerable care is needed to ensure that environmental conditions will not violate assumptions underlying the methodology (see Kaimal and Finnigan, 1994, Atmospheric Boundary Layer Flows, Oxford Press, N.Y.). Instruments are typically deployed from fixed platforms (e.g. towers or scaffolding) meters to 10's of meters above the surface. Depending upon instrument height above the surface, eddy covariance measurements are representative of a surface-atmosphere flux occurring over an upwind area on the order of hectares to 100's of hectares. Usually the higher the instruments are placed, the larger the upwind area sampled. In some instances, eddy covariance instruments have been flown aboard aircraft, enabling researchers to make comparative flux measurements over 100's of kilometers.


For More Information

Dean E. Anderson
USGS, WRD
MS413, Box 25046, DFC
Denver, CO 80225
303-236-5691
deander@usgs.gov
           Dave Stannard
USGS, WRD
MS413, Box 25046, DFC
Denver, CO 80225
303-236-4983
distanna@usgs.gov


Return to USGS Carbon Cycle Research

Accessibility FOIA Privacy Policies and Notices


Take Pride in America home page. USA.gov logo U.S. Department of the Interior | U.S. Geological Survey
URL: http://geochange.er.usgs.gov/carbon/ghg/
Page Contact: ESD Webserver Team
Page Last modified: Tue 2-May-2006 14:11:24 MDT