Project Overview

The Megacities Carbon Project will develop and test methods for monitoring the greenhouse gas emissions of the largest human contributors to climate change: cities and their power plants. Pilot activities have already begun in the megacities of Los Angeles and Paris that build on existing research infrastructure there and collaborations between the teams involved. Discussions are also underway regarding inclusion of a third sister city in Sao Paulo, Brazil. Our goal is to demonstrate a scientifically robust capability to measure multi-year emission trends of carbon dioxide (CO2), methane (CH4), and carbon monoxide (CO) attributed to individual megacities and selected major sectors. After the three-city pilot is established the project may be expanded to include additional cities. A key element of the project will be open and transparent data sharing between the international partners.

The Los Angeles component of the project is jointly funded by the US National Institute of Standards and Technology (NIST), National Aeronautics and Space Administration (NASA), National Oceanographic and Atmospheric Administration (NOAA), and the Keck Institute for Space Studies (KISS). In-kind contributions provided by the California Air Resources Board and UC Discovery.

Project Vision

Although methodological studies have been conducted in small- to medium-size cities with stable emissions, there has not yet been an effort to extend these techniques to the more complex and representative environments in megacities - and ultimately, towards a global urban monitoring framework. To address these gap, the Megacities Carbon Project is being established in Los Angeles – with partnerships in other global megacities - that leverages and extends established measurement infrastructure in those cities and applies the techniques being developed by the smaller methodological studies.

Ultimately, sustained monitoring of the atmospheric domes of carbon dioxide, methane, and carbon monoxide for the world’s cities, scientifically robust analysis linking that information to emission activity, and transparent data sharing will provide decision makers with information critical to assessing the ultimate efficacy of emission mitigation policies. It will also help local leaders who are taking action now explain to their communities and stakeholders the importance of and progress towards addressing the single biggest human contribution to climate change: carbon emissions from cities

Project Objectives

  1. Develop and demonstrate measurements systems capable of quantifying trends in the anthropogenic carbon emissions of the Los Angeles Megacity (target: 10% change in Fossil Fuel CO2 over 5 years)
  2. Reduce uncertainty in CH4 emissions to the same level reported for CO2 and CO emissions.
  3. Validate assertions (test hypotheses) that "current uncertainties in urban CO2 and CO inventories are < X%".
  4. Attribute observed emission signatures to key sectors and activity in the megacity.
  5. Identify and quantify CH4 emission point sources.
  6. Advance measurement capabilities, models, and protocols to bridge the research to operations gap.
  7. Extend methods and techniques to many cities and establish open and transparent sharing of data and analysis and measurement methods (towards capacity building).


Why Are Cities Important?
Cities and their power plants are the largest sources of greenhouse gas emissions from human activity. As of 2010, urbanization has concentrated more than half of the world's population, at least 70% of fossil fuel carbon dioxide emissions, and a significant amount of anthropogenic methane into a small fraction of the Earth’s land surface. Currently, the 40 or 50 largest cities together represent the third largest emitter of fossil fuel CO2 after China and the US.

Cities are changing faster than countries
Greenhouse gas emissions are changing rapidly in cities both in terms of mitigation or stabilization as well as unconstrained growth.

Some of the world's largest cities are leading the way with aggressive greenhouse gas stabilization policies. For example, the Green LA Plan (2007) calls for 35% reduction (vs 1990 levels) by 2030 and the Paris Climate Plan (2007) 25% (vs 2004) by 2020. These and other megacities (population of at least 10 million people) form the core of collaborations such the Climate 40 (C40) group of mayors.

Meanwhile explosive, unconstrained growth is occurring in other cities. Global urbanization is projected to double by 2050. Some megacities in the developing world are currently undergoing population growth of 4%/year and emissions growth of 10%/year. The number of megacities is expected to grow from 23 to 37 between 2010 and 2025.

Current uncertainties in urban emissions presents a risk
Currently, estimates of greenhouse gas emissions from many cities are either not available or are generated using "bottom-up" accounting methods. Where such estimates are available agencies collect data from different sectors and use emission-factors to calculate the emissions associated with a given activity. The results are then tabulated in an emission inventory for the city (typically annually). Inventories are critical for managing greenhouse gas stabilization policies but they carry a risk for decision makers in terms of uncertainty. The degree of uncertainty in these emission inventories depends on the quality and completeness of the emission activity data, accuracy of the emission factors, and the estimation, quality control, error quantification, and verification processes applied to them. Hence "your mileage may vary" significantly by sector, gas, and city in terms of the resulting uncertainty. In some cases, differences of 50% or more has been observed when comparing inventory estimates from atmospheric measurements for a specific location, sector or gas. Additionally, since most stabilization policies are concerned with multi-year trends in emissions uncertainties of a few % per year can translate in errors over several years equal to or larger than the expected change in emissions. These uncertainties and errors represent a risk to decision makers in terms of drawing the wrong conclusions when trying to answer the following questions: are policies having the intended impact and if not, why and how should they be change? are the policies being implemented cost-effective or could they be made more efficient?

Reducing uncertainty is the primary motivation for jointly applying improved inventories and atmospheric observations to monitor trends in emissions. By applying independent and accurate measurements we seek to identify errors in data or assumption to help improve the fidelity of the inventories and ultimately, to validate what is actually being emitted into the atmosphere.

Establishing baselines and monitoring trends
Currently there are no established baseline estimates of the atmospheric emission signatures of large cities. Without accurate baselines, future estimation of trends carries a large degree of uncertainty. There is an urgent need to establish these baselines given the rapid change in emissions of the largest cities projected over the next decade.

Additionally, no current or planned system is capable of monitoring the atmospheric trends of carbon attributed to the world’s largest cities despite recent field experiments demonstrating scientifically robust methods for assessing carbon emissions and rapid improvements in measurement technology. A concerted effort is needed to transform emerging scientific methods and technologies into an operational monitoring system to support urban carbon management decisions.

Scientific Challenges

While our vision is to ultimately establish a global, urban carbon monitoring system that addresses the largest human contribution to climate change there are still many scientific and technical challenges to overcome. The Megacities Carbon Project is conducting pilot studies in Los Angeles, Paris and other megacities to help leverage and extend recent developments in measurement science to effectively bridge the gap between basic research and, ultimately, an operational monitoring system. We are coordinating with related efforts in smaller cities such as Indianapolis and Boston to address the following challenges:

  • Disentangling the relative contributions of fossil-fuel use and biosphere (photosynthesis and respiration) to CO2 concentration in the atmosphere (and other confusion issues with methane) requires measuring multiple gas species and radioisotopes as well as higher space-time resolution for observations and models.
  • Accounting for the effects of a moving atmosphere (winds and boundary layer depth) is critical for properly translating gas concentration measurements into emission estimates.
  • Linking the "top down" emission estimates derived from atmospheric data with space-time resolved "bottom up" emission estimates derived from energy activity data is required to provide policy-relevant data products.

Figure 1
Gridded annual fossil fuel CO2 emissions from a medium-size city (Indianapolis) show distinct gradients at different spatial scales. Right: CDIAC 2006 emissions for the CONUS plotted on a 1° (~100 km) show avg flux 200-600 gC/m2/yr. Middle: Vulcan 2002 emissions for the ~10,000 km2 area centered on Indianapolis on a 10 km grid. Left: Hestia 2002 emissions for the urban core on a 1 km grid. The Vulcan and Hestia plots use log-normal scales (typically >20,000 gC/m2/yr).

Methods and Project Elements

Tower Network

A network of 13 surface measurement sites within and around the Los Angeles basin provides the backbone of the Megacities Carbon Project monitoring system. These sites are equipped with in-situ greenhouse gas analyzers that continuously sample the air drawn by pumps through air hoses mounted on radio towers and tall buildings. All of the sites measure CO2, all but one measure methane (CH4), and about half currently measure Carbon Monoxide (CO). Some of the sites have been operating for over 10 years but the main installation effort started in late 2012. The current network installation was complete in August 2015. The site locations were carefully selected to ensure the network provides complete coverage of the basin and to avoid “contamination” of known nearby pollution sources. In addition to the sites located within the basin focused on measuring urban air, there are 3 sites outside the LA basin (Victorville, San Clemente Island and La Jolla) that provide a relatively clean “background” reference measurement as well as (sometimes, when the winds shift) measurements of outflowing air from the basin. The network is currently undergoing commissioning and checkout. See our network page for information about specific sites.


The California Laboratory for Atmospheric Remote Sensing (CLARS) is a facility operated by JPL on Mount Wilson. At an altitude of 1670 meters (about 5500 feet) above sea-level, CLARS has a near panoramic view of the LA basin. Given it’s vantage point (essentially a low altitude geostationary satellite), CLARS is used to test prototype instruments designed to ultimately fly in space. It has the advantage of being able to scan most of the LA basin several times of day and product “images” of CO2 and methane concentrations – similar to a CAT scan. See our multimedia for more info.


The Total Carbon Column Observating Network (TCCON, pronounced “Tee-Con”), was developed to help validate the observations of CO2 and other greenhouse gases from satellites like NASA’s Orbiting Carbon Observatory and JAXA’s GOSAT missions. It consists of a network of ground-based Fourier Transform Spectrometers that measure solar spectra in the near-infrared spectral region. TCCON sites track the sun during daylight hours and essentially sample the same atmospheric column as the satellites overhead except only in one direction (whereas satellites have a two way path since they use sunlight reflected off the earth’s surface). For LA, we have two TCCON sites: one located at Caltech in Pasadena and one located in the high desert at NASA’s Armstrong Flight Research Center (AFRC, formerly known as Dryden Flight Research Center). These two cites provide an urban and rural site, respectively, for tracking the “urban dome” of CO2 and methane over LA. See Caltech's TCCON site for more information about the TCCON network.


The first “A” in NASA stands for Aeronautics and the agency has a large suite of remote-sensing instruments designed to operate on aircraft (most of which were designed as prototypes for future satellite instruments). Some of these airborne instruments are capable of studying greenhouse gases. Imaging spectrometers are a class of instrument that is particularly relevant to understanding greenhouse gas emissions in complex urban environments. JPL’s Airborne Visible/InfraRed Imaging Spectrometer (AVIRIS) and Hyperspectral Imaging Spectrometer (HyTES) are both capable of surveying large areas like the LA basin to detect and produce images of invisible plumes of methane gas from individual point sources. While these aircraft don’t fly regularly over LA we periodically conduct airborne campaigns to study methane point sources such as landfills, waste water treatment facilities, refineries, oil fields, natural gas facilities, dairies and other sources of methane emissions in the LA basin. See here for more information about AVIRIS and HyTES.


The gas analyzers of the LA tower network provides continuous, 24/7 monitoring of CO2 and (at selected sites) CH4 and CO. That data is recorded locally and transmitted daily to the Megacities data portal. However three of these sites are also equipped with Programmable Flask Packages (PFPs) that automatically collect small samples of air (typically twice per week) and store it in small bottles or flask. Those flasks are then collected periodically and shipped to NOAA’s Earth System Research Laboratory in Boulder, Colorado where they are analyzed for many gases including a radioisotope of CO2 (14CO2). This information is used to “calibrate” the relationship between the continuous measurements of CO2 and CO and the relative contributions of fossil fuel combustion and the biosphere (plants) to CO2 in the atmosphere. 14CO2 is useful in this process because the carbon stored in fossil fuels like coal, oil and natural gas, has been stored underground for a long time and most of the 14CO2 radio-isotope has decayed compared to newer carbon stored in plants. So 14CO2 help provides a way to “fingerprint” the relative contributions from fossil fuel combustion and plants to the CO2 in the atmosphere over LA.


NASA’s Orbiting Carbon Observatory (OCO-2) and JAXA’s GOSAT satellite are examples of space-based measurements of greenhouse gases over LA and other parts of the world. OCO-2 only measures CO2 but does so with unprecedented precision and spatial resolution. GOSAT has coarser resolution but measures both CO2 and CH4. Satellite observations complement surface monitoring systems by covering very large areas quickly. A major objective of the Megacities project is to serve as a “testbed” for these and other satellites so scientists can explore ways to use the satellite data to study greenhouse gas emissions from cities around the world.

Surface Intensives

In addition to sustained monitoring from the surface measurement network and satellites, periodic intensive measurement campaigns are conducted to address specific questions (e.g., the 2010 CALNEX campaign).

Some examples of intensive campaigns for Los Angeles include:
An initial campaign was conducted in Fall 2012 to characterize Planetary Boundary Layer Height (PBLH) in the Los Angeles basin. PBLH, along with winds, has a major influence on atmospheric concentrations of greenhouse gases and can lead to misinterpretation of greenhouse gas measurements. Past studies of this topic in LA have tended to focus on air quality issues which are related but distinct from the challenge of greenhouse gas measurements. While some PBLH information can be derived from existing radar wind profilers at the major airports across the basin their relatively coarse vertical resolution complicates efforts to test the higher resolution models. This campaign had three primary objectives: 1) measure trends in PBLH at several representative sites across the basin with vertical resolutions of better than 20 meters, 2) measure spatial gradients in PBLH across the basin, and 3) evaluate different PBLH measurement technologies to informing decisions about the optimal monitoring network for the LA basin. The initial phase of the intensive involved sustained observations at Caltech and USC with several days of rawinsonde launches.

In Spring and Summer 2014 we coordinated surface observations with aircraft campaigns involving NASA's Hyperspectral Thermal Emission Spectrometer (HyTES). These campaigns were focused on characterizing point sources of methanes from a variety of sources including natural gas infrastructure, oil production, landfills, wastewater treatment, natural seeps and dairies - all of which occur in LA.

In fall 2014 we conducted intensive campaign to collect whole air flask samples across the LA basin at high frequency. This began a 3 year NASA and NOAA activity to characterize the spatial and temporal distribution of fossil fuel CO2 emissions across the LA megacity domain using measurements of 14C, CO2 and CO from surface measurements and CO2 and CO from satellite observations.

Modeling and Analysis

Nearly all of the observing systems described here measure concentrations or “mixing ratios” of CO2 and other greenhouse gases in the atmosphere. However those measurements alone only provide information about how much of a particular gas is present at the location and time of the measurement. To answer the questions that many scientists and decision makers are more interested in - understanding the actual location and strength of the many sources of greenhouse emissions in and around the LA basin – we need to combine those greenhouse gas concentration measurements with computer models of how the atmosphere moves. The meteorology of LA, particularly wind and how the surface layer of the atmosphere moves up and down during the say, is complex and often dominates our efforts to estimate greenhouse gas emissions. A major focus of the Megacities project is combining cutting-edge models of atmospheric circulation and the greenhouse gas concentration data to study emission fluxes.

A critical third component of this system is a special high resolution data set called “Hestia” that provides the best-available prior estimate of fossil fuel CO2 emissions over Los Angeles County and the 4 adjacent counties. Hestia represents fossil fuel CO2 emissions for all of the major economic sectors in the LA basin including residential, commercial, industrial, onroad, etc. Hestia provides this information at roughly 1 km (sometimes finer) and hourly resolution. By “driving” our modeling system with the prior Hestia emissions we can simulate concentrations of CO2 in the atmosphere and compare them with our CO2 measurements around the basin. We can also use the CO2 measurements and the model to “invert” for the emissions and then compare with the Hestia output and then adjusting things until they match. This process of optimization and comparison is a synthesis of techniques and it ultimately allows us to improve our understanding of greenhouse gas emissions in LA – both for specific locations and key sectors.