Inverse modeling of CO2 sources and sinks using satellite data: a synthetic inter-comparison of measurement techniques and their performance as a function of space and time

[摘要] Currently two polar orbiting satellite instruments measure CO₂concentrations in the Earth's atmosphere, while other missions are plannedfor the coming years. In the future such instruments might become powerfultools for monitoring changes in the atmospheric CO₂ abundance and toimprove our quantitative understanding of the leading processes controllingthis. At the moment, however, we are still in an exploratory phase wherefirst experiences are collected and promising new space-based measurementconcepts are investigated. This study assesses the potential of some of theseconcepts to improve CO₂ source and sink estimates obtained frominverse modelling. For this purpose the performance of existing and plannedsatellite instruments is quantified by synthetic simulations of their abilityto reduce the uncertainty of the current source and sink estimates incomparison with the existing ground-based network of sampling sites. Our highresolution inversion of sources and sinks (at8°x10°) allows us to investigate the variation ofinstrument performance in space and time and at various temporal and spatialscales. The results of our synthetic tests clearly indicate that thesatellite performance increases with increasing sensitivity of the instrumentto CO₂ near the Earth's surface, favoring the near infra-redtechnique. Thermal infrared instruments, on the contrary, reach a betterglobal coverage, because the performance in the near infrared is reduced overthe oceans owing to a low surface albedo. Near infra-red sounders cancompensate for this by measuring in sun-glint, which will allow accuratemeasurements over the oceans, at the cost, however, of a lower measurementdensity. Overall, the sun-glint pointing near infrared instrument is the mostpromising concept of those tested. We show that the ability of satelliteinstruments to resolve fluxes at smaller temporal and spatial scales is alsorelated to surface sensitivity. All the satellite instruments performedrelatively well over the continents resulting mainly from the larger priorflux uncertainties over land than over the oceans. In addition, the surfacenetworks are rather sparse over land increasing the additional benefit ofsatellite measurements there. Globally, challenging satellite instrumentprecisions are needed to compete with the current surface network(about 1ppm for weekly and 8°x10° averaged SCIAMACHYcolumns). Regionally, however, these requirements relax considerably,increasing to 5ppm for SCIAMACHY over tropical continents. Thispoints not only to an interesting research area using SCIAMACHY data, butalso to the fact that satellite requirements should not be quantified by onlya single number. The applicability of our synthetic results to real satelliteinstruments is limited by rather crude representations of instrument and dataretrieval related uncertainties. This should receive high priority in futurework.

[学科分类] 大气科学