Go to Content | Go to Page Updated Information | Go to Footer
Welcome to the California Energy Commission
Public Interest Energy Research Program: Final Project Report

cover of report Development of an Implementation Plan for Atmospheric Carbon Monitoring in California

Publication Number: CEC-500-2005-123
Publication Date: August 2005
PIER Program Area: Energy-Related Environmental Research

The executive summary, abstract and table of contents for this report are available below. This publication is available as an Adobe Acrobat Portable Document Format Files. In order to download, read and print PDF files, you will need a copy of the free Acrobat Reader software installed in and configured for your computer. The software can be downloaded from Adobe Systems Incorporated's website.

Download Report in Acrobat PDF ( 43 pages, 1.2 megabytes - note size! )


Executive Summary

Introduction

Carbon dioxide (CO2) is a greenhouse gas (GHG) that is emitted as the result of combusting fossil fuels for electricity production, transportation, industry, and other uses. Worldwide, roughly half of the CO2 emitted by fossil fuel combustion remains in the atmosphere, leading to sustained increases in atmospheric CO2 concentrations. In-state electricity generation accounts for about 16% of California’s anthropogenic CO2 emissions.[ 1 ] Because these emissions contribute to human-induced climate change, the California Energy Commission (Energy Commission) has identified the study of CO2 and other GHGs as a priority area for research. Future cost and production strategies for electricity generation may be affected by economic and policy responses to global warming. Therefore, it will be important to determine the contribution of fossil fuel combustion and other sources of CO2 exchange to atmospheric CO2 levels.

Currently, the Energy Commission estimates CO2 emissions using accounting data for fossil fuel emissions and a combination of inventory surveys and simple models for net ecosystem exchange (NEE)[ 2 ] of CO2. The net ecosystem fluxes occur continuously and uncertainties associated with these fluxes are poorly quantified.

At the national level, the carbon cycle community was posed with the analogous problem of providing defensible estimates of net CO2 (and other GHG) exchange. In response, the research community suggested a North American Carbon Program (NACP) to provide a robust approach for reducing the uncertainties in regional CO2 exchange estimates. The NACP is envisioned to combine inventory methods of biosphere carbon stocks and fossil fuel emissions, direct CO2 flux measurements,[ 3 ] and measurements of atmospheric CO2 concentrations, all coupled within an inverse analysis[ 4 ] employing ecosystem-atmosphere models of NEE and atmospheric transport.

Measurement of CO2 concentration gradients on regional scales is challenging because the concentration differences generated by NEE are in the range of < 1 part per million (ppm) to approximately 20 ppm. Sufficiently accurate measurements can be made for signals greater than about 1 ppm from the instrument perspective, because the state-of-the-art in measurement accuracy is approximately 0.2 ppm of CO2. A careful examination of the expected concentration signals from NEE and fossil fuel CO2 emissions is an important first step in ensuring that atmospheric methods can be applied successfully to California.


Purpose

Design a system of atmospheric concentration measurements as part of a larger strategy for monitoring CO2 exchange in California.


Project Objectives

The LBNL researchers identified four objectives:

  1. Predict temporal and spatial distributions of NEE and meteorology in the Western United States for representative time periods covering the seasonal cycle of NEE.

  2. Predict temporal and spatial distributions of fossil fuel CO2 emissions in the Western United States.

  3. Predict temporal and spatial distributions of CO2 concentration signals due to NEE and fossil fuel CO2 emissions inside and outside California.

  4. Use the atmospheric CO2 concentration signals predicted for existing and potential future monitoring stations to judge the suitability of these stations for quantifying NEE within California.


Key Results

  • The predicted atmospheric CO2 concentration signals from NEE at many locations within California are both measurable with existing methods and often comparable or significantly larger than signals from fossil fuel emissions.

  • The predicted CO2 concentrations at marine buoys and coastal stations are significantly affected by terrestrial signals and cannot be used to represent global background CO2 concentrations without thoughtful correction.

  • A combination of existing terrestrial sites (including flux measurement sites and tall communication towers) would likely be effective as observation stations for separating CO2 concentration signals from NEE within California from signals attributable to fossil fuel emissions and NEE outside California.

  • Identification of a protocol for measuring gas concentrations that contribute to the estimation of net ecosystem exchange, with regional specificity and in the presence of combustion-related GHGs.


Project Recommendations

CO2 Observing Stations, Sampling Frequencies, and Measurement Protocol

  • New terrestrial sites sensitive to California NEE: New terrestrial observation stations should be deployed, likely on existing communications tall towers in California where signals are predominantly influenced by NEE or fossil fuel CO2 emitted within California.

  • Sampling frequency: Automated air sampling should be conducted continuously, to accurately capture the concentration during the well-mixed afternoon period and the diurnal variation. Data should be collected and averaged into approximately 30 minute intervals for further analysis. However, if resources are limited, periodic sampling with National Oceanic and Atmospheric Administration’s (NOAA) flask sampling systems[ 5 ] could capture day-versus-night differences, which would provide valuable initial data to test the model predictions from this and other studies.

  • Measurement protocol: Air sampling and analysis, instrument design, calibration, and analysis should follow established methods to ensure accuracy and minimize systematic errors. Continuous measurements should be checked against measurements with NOAA flask systems. In addition to the CO2 data, standard ancillary meteorological data (e.g., temperature, relative humidity, wind velocity) should also be gathered and recorded.

  • Additional stations: Observations are planned for a variety of platforms to monitor background air entering California for carbon cycle and air quality purposes. These observations will likely include the three planned marine stations planned by NOAA, periodic aircraft profiling planned by NOAA, and sampling from one or more mountaintop stations. It is also expected that existing flux measurement sites will be augmented with accurate and precise CO2 concentration measurements. These data sets should be included in future analyses.

Uncertainty Analysis and Station Optimization

  • Future work should expand on the study described in this report to predict the uncertainty that will likely be obtained in estimating the total NEE from inside California from the measured concentration data.

  • Each of the major sources of uncertainty (e.g., spatial coverage of stations, background subtraction, and transport error) should be quantified in terms of uncertainty in estimating NEE.

  • Future simulations should include several complete annual cycles to sample different weather patterns and to capture interannual variations in NEE.

Measurement of Additional Trace Gas Species

  • Where possible, carbon monoxide (CO) measurements should be included with CO2 measurements to quantify fossil fuel CO2 emissions. The use of other sources of CO (and other pollutant) data from the California Air Resources Board (CARB) might play a valuable part in this effort, particularly near urban areas where pollutant concentrations are high enough to be clearly detected with available instruments.

  • Because fossil fuel CO2 contains no radiocarbon (14C), measurements of 14CO2 in atmospheric samples can be used to infer the fraction of CO2 due to fossil fuel combustion. Because of the high cost for precision analysis, this technique can probably only be applied with a low sampling frequency.

  • The use of radon gas (222Rn) as a tracer of atmospheric mixing should be explored for use in combination with CO2 measurements to quantify net regional CO2. This technique may be well suited to the problem of California carbon budgets, because much of the air entering California is of marine origin and will be depleted in 222Rn.

Data Synthesis

  • A crucial aspect of monitoring California’s CO2 emissions using the concentration measurements in an inverse or “top-down” approach will be to sensibly incorporate “bottom-up” information (e.g., ecosystem production from forest inventories and crop yields, fossil CO2 emissions from improved inventory methods, eddy covariance CO2 flux data from towers and aircraft flights, and ecosystem models of CO2 exchange) . This approach is an emerging area and developments in the North American Carbon Program (NACP) will be valuable to addressing these issues in California.

Estimating Non-CO2 GHG Emissions

  • The methods that the LBNL researchers have applied here could be used to develop a measurement strategy for non-CO2 GHGs. Because the relative uncertainties for fluxes of non-CO2 GHGs are considerably larger than for CO2, atmospheric budget and inverse approaches may be particularly effective in significantly reducing those uncertainties if atmospheric observations can be performed.


Footnotes

  1. California Energy Commission. November 2001. Inventory of California Greenhouse Gas Emissions and Sinks: 1990–1999. P600-02-001F. Figure ES-8.


  2. NEE is the net flow of CO2 from terrestrial ecosystems to the atmosphere. For both NEE and fossil fuel CO2 emissions, fluxes are defined so that a positive flux causes an increase in atmospheric CO2 concentrations.


  3. Direct flux measurements quantify CO2 exchange over a limited area using micrometeorological methods. For example, the eddy covariance method employs rapid measurements of vertical wind velocity and CO2 concentration to calculate the net flux of CO2 as average of CO2 concentration weighted by vertical velocity. Equivalently, the flux can be viewed as the net difference in the amount of CO2 carried upward minus downward with the atmospheric eddies.


  4. Mathematical inverse methods estimate the spatial and temporal distribution of surface fluxes that are most consistent with measurements of CO2 concentration by working backward from concentration to flux under the assumption that CO2 concentrations are the result of surface fluxes being transported in the atmosphere by meteorology. See for example, Kasibhatla, P., and American Geophysical Union. 2000. Inverse methods in global biogeochemical cycles. Washington, D.C.: American Geophysical Union.


  5. NOAA conducts routine sampling and laboratory analysis of air samples using standardized methods. See www.cmdl.noaa.gov/ccgg/flask.html.




Abstract

This report describes the design of atmospheric CO2 concentration measurements that, in combination with other measurements and models, would be used to quantify regionally distributed CO2 exchanges from California's terrestrial ecosystems and CO2 emissions from fossil fuel combustion. Using models of net ecosystem CO2 exchange (NEE), fossil fuel CO2 emissions, and regional meteorology, the LBNL research team predicted CO2 concentration “signals” in the atmosphere. Researchers subdivided the CO2 sources into four components: (1) NEE inside California, (2) NEE outside California, (3) fossil fuel CO2 inside California, and (4) fossil fuel CO2 outside California. Maps of predicted atmospheric CO2 concentration signals from these four sources largely mirror the instantaneous emissions near strong sources, but plumes of CO2 enriched or depleted air are predicted to advect far from their sources. Using the predicted concentration signals, LBNL identified a baseline set of observing stations from existing and possible future sites that could be used to characterize in-state and out-of-state ecosystem and fossil fuel contributions to atmospheric CO2 concentrations. The research team also identified a measurement protocol for making concentration measurements at the stations. Although beyond the scope of this project, future work should estimate the uncertainties in estimating California's NEE that would be obtained using atmospheric concentration data from the stations identified herein.

Keywords: greenhouse gas emissions, atmospheric monitoring, inverse methods, land-surface CO2 modeling



Table of Contents

Preface ii

Abstract vi

Executive Summary 1

1.0 Introduction 6

1.1. Background and Overview 6

1.2 Project Goal and Objectives 8

1.3 Report Organization 8

2.0 Project Approach 9

2.1. Prediction of NEE and Meteorology 9

2.2. Prediction of Fossil Fuel CO2 Emissions 11

2.3. Prediction of Concentration Signals 12

2.4. Identification and Evaluation of Observation Stations 13

3.0 Results and Discussion 13

3.1. Spatiotemporal Distribution of Net Ecosystem Exchange 13

3.2. Spatiotemporal Distribution of Fossil Fuel CO2 Emissions 15

3.3. Spatiotemporal Distribution of CO2 Concentrations in the Atmosphere 16

3.4. Identification and Evaluation of Observation Stations 19

3.4.1. Existing and Potential Future Stations 19

3.4.2. CO2 concentration signals at monitoring station locations 20

3.4.3. How Well Do Concentration Signals at Monitoring Sites Allow Discrimination of NEE Inside California from Other Signals? 22

4.0 Conclusions and Recommendations 24

4.1. Observation Stations, Sampling Frequency, and Measurement Protocol 24

4.2. Uncertainty Analysis and Site Optimization 25

4.3. Additional Trace Gas Species for CO2 Source Attribution 26

4.4. Synthesis of Bottom-Up and Top-Down Methods 27

4.5. Potential for Monitoring Non-CO2 GHG Emissions 27

5.0 References 28

6.0 Glossary 31

Appendix A. Diurnal and Seasonal Variations in NEE A-1

Appendix B. Seasonal Cycle of Mean Midday CO2 Concentration Signals B-1

Appendix C. Seasonal Cycles of Mean Midday Concentration Signals at Observing Stations C-1


List of Figures

Figure 1. Vegetation distribution used to characterize land cover in the Western United States for the simulations described in this report. 11

Figure 2. Maps showing the diurnal cycle of predicted net ecosystem exchange, NEE (µmol CO2 m-2 s-1), for a day in June. 14

Figure 3. Maps showing the diurnal cycle of estimated fossil fuel CO2 emissions for the Western United States for a weekday in June. 15

Figure 4. Maps showing diurnal variation in surface layer CO2 concentrations resulting from NEE for the day shown in Figure 2. 16

Figure 5. Maps of diurnal variation in surface layer CO2 concentrations resulting from fossil fuel CO2 emission for June time period as in Figure 4. 17

Figure 6. Maps of June midday mean surface layer CO2 concentrations resulting from NEE inside (upper left) and outside (upper right) California, and fossil CO2 emissions inside (lower left) and outside (lower right) California. 18

Figure 7. The positions of existing and possible future observations stations examined in this study. 19

Figure 8. Mean CO2 concentration signals from NEE in and out and fossil fuel CO2 emissions in and out of California for each of the four months examined in this study. 21

Figure 9. Standard deviation of midday CO2 signals from NEE inside California (top left), showing predicted variation in daytime signals for each season. 23

Figure A1. Diurnal variation in June NEE A-1

Figure A2. Seasonal variation in daily mean NEE A-2

Figure B1. Signals in March B-1

Figure B2. Sinals in June B-1

Figure B3. Signals in September B-1

Figure B4. Signals in December B-1

Figure C1. Seasonal cycles of mean midday concentration CO2 signal for observing stations C-1


List of Tables

Table 1. Existing and potential monitoring stations. 20


Page Updated:
Go to Content | Go to Page Updated Information | Go to Footer