Sensitivity studies for space-based laser measurements of atmospheric CO2 concentration towards future NASA mission ASCENDS

Global measurement of atmospheric CO2 concentration from space is greatly desired for global and regional carbon budget studies. The biggest challenge for CO2 remote sensing from space is to achieve the high-precision science measurement requirement (~ 1 ppmv or 0.3% on regional scales) so that the satellite measurements will be able to remarkably reduce current large uncertainties about carbon sources and sinks (Tans et al., 1990; Fan et al., 1998). CO2 shortwave infrared bands at 1.57, 1.6 and 2.1 m have been studied and shown to possess great potential to achieve such high-precision measurement (O’Brien and Rayner, 2002; Kuang et al., 2002; Dufour and Breon, 2003; Mao and Kawa, 2004). The Orbiting Carbon Observatory (OCO), the NASA Earth System Science Pathfinder Project (ESSP) mission, and the Japanese Greenhouse gases Observation SATellite (GOSAT) have adopted these bands to measure CO2 from space using surface reflected sunlight (Crisp et al., 2004; Yokota et al., 2004). However, using these weak vibration-rotation bands requires line-resolving spectral resolutions to achieve measurement precision (Dufour and Breon, 2003; Mao and Kawa, 2004; Fu et al., 2008). In other hand, a major issue in the passive approach using reflected sunlight with these bands is the contamination of aerosol and/or cirrus cloud scattering in the sunlight path (Mao and Kawa, 2004; Aben et al., 2007). Scattering by the particles in the atmosphere will modify the sunlight path length and thus change total column CO2 absorption and then result in retrieval errors greater than measurement precision target. This type of errors varies in time and space, depending on solar illumination and satellite viewing geometry, surface reflectivity and aerosols/clouds physical (i.e., altitude) and optical properties (i.e., scattering phase function). The National Research Council of the U.S. National Academies has recommended a future mission called Active Sensing of CO2 Emissions over Nights, Days, and Seasons (ASCENDS) in its Decadal Survey Earth Science and Applications from Space: National Imperatives for the Next Decade and Beyond. This mission aims “to produce global atmospheric column CO2 measurements without seasonal, latitudinal, or diurnal bias using simultaneous laser remote sensing of CO2 and O2.” NASA Goddard Space Flight Center is developing an active approach as a candidate toward this mission. A series of sensitivity studies have been conducted for this development. Following is a summary of the sensitivity study results and more details will be presented during the meeting. (1). Measurement Strategy The laser technique for this approach uses the 1.57 m CO2 absorption band and samples several wavelengths on one strong absorption line in this band (Abshire et al., 2008). It measures differential absorption at local nadir mode and uses the ratio of the on-line to off-line signal to retrieve CO2 amount without aerosol and thin cloud absorption terms. The lasers in this approach are pulsed and the surface return signal can be well separated from that returned by the atmosphere using time gating in the receiver. Thus, compared to passive approaches and other active laser approaches (Menzies and Tratt, 2003; Ehret et al., 2008; Browell et al., 2008; Heaps et al., 2008), the aforementioned influences from atmospheric scattering can be greatly reduced by this method. Meanwhile, on-board O2 measurement using O2 A-band at 0.76 m is recommended using the same method and technology for dry air abundance in the same atmospheric path in order to retrieve CO2 concentration (mixing ratio) with regard to dry air. The time-gating and pulsed laser approach will yield even greater benefits in the O2 A-band where atmospheric scattering is larger and was a major error source in path length and surface pressure measurements (e.g., Mitchell and O’Brien, 1987). (2). Line and Wavelengths Selections Sensitivity studies showed the optimal CO2 line in the 1.57 m band was the strongest CO2 line in the R-branch centered at 1572.335 nm. This line has overall the minimum sensitivity to atmospheric temperature change, particularly to the lower atmospheric temperature change, and is close to the available laser which is centered at 1530 nm. Absorption measurements on several wavelengths on this line provide vertical weighting functions peaking at desired layers. Measurement at wavelength with approximately 50% absorption targets CO2 variation in the atmospheric boundary layer for sources and sinks estimate. Measurements at wavelengths closer to the line center have greater absorption, providing free troposphere CO2 information related to vertical and horizontal CO2 transport, equivalent to CO2 retrievals from advanced temperature sounders such as AIRS and IASI. Measurement at line center can yield stratospheric CO2 which is about 4-5 years older than tropospheric CO2. The optimal line for the on-board O2 measurement is located in the middle of P-branch of O2 A-band and centered at 764.7 nm with the minimum sensitivity to atmospheric temperature variation. The wavelength in the ridge of the doublet line is ideal for O2 measurement from space for the least temperature sensitivity, vertical weighting function peaking at the surface and the smallest impact of potential wavelength drift in the fly. (3). Ancillary Data Requirements Atmospheric temperature fluctuations can modify line absorption measurements and change CO2 mixing ratio as well even without any CO2 source/sink changes. High spatial and temporal resolution atmospheric temperature profiles are thus required to accurately derive CO2 concentration from O2/CO2 absorption measurements. Advanced temperature sounders, AIRS, IASI and future NPOESS/CrIS can provide mesoscale temperature profiles within a precision better than 1 K in 1-km-thick layers (Susskind et al., 1998). Temperature analyses from weather forecast models can provide similar precision of temperature data also to meet high-precision CO2 retrievals. Atmospheric water vapor is another ancillary data required for CO2 mixing ratio calculation. Besides minor water vapor absorption in the CO2 line used in this development, water vapor exerts a partial pressure cause line broadening and an error in the retrievals. High spatial and temporal resolution atmospheric water vapor profiles from advanced sounders like AIRS or from weather forecast models are also needed in deriving CO2 concentration particularly in the tropics and other moisture atmospheres. (4). Laser Requirements and other Measurement Optimizations For a given sensor design the surface returned pulse strength and hence signal-to-noise ratio (SNR) depends strongly on laser peak power, orbit altitude, the atmospheric transmission and surface reflectivity. For an example, SNRs at both off-line and on-line are a function of laser peak power. The relative error of retrievals inferred from on-line to off-line ratio decreases to less than 0.003 or 0.3% (measurement precision target) when laser peak power increases to 3000 watts or greater. Meanwhile, lower orbit altitude, higher surface reflectivity and atmospheric transmission will enhance SNR and reduce retrieval errors.