气候变化领域集成服务门户

We found success with a relatively simple method based on calibration and checking with surface monitors and a characterization of vertical mixing, and incorporating specific understanding of the region's climatology. We estimate PM2.5 to within ĝˆ1/47 μg m-3 root mean square error (RMSE) and with R values of ĝˆ1/40.9, based on remotely sensed multi-angle implementation of atmospheric correction (MAIAC) observations, and certain further work will improve that accuracy. Mapping is at 1 km resolution. This allows a time sequence of mapped aerosols at 1 km for cloud-free days. We describe our technique as a "static estimation." Estimation procedures like this one, not dependent on well-mapped source strengths or on transport error, should help full source-driven simulations by deconstructing processes. They also provide a rapid method to create a long-term climatology.

Essential features of the technique are (a) daily calibration of the AOT to PM2.5 using available surface monitors, and (b) characterization of mixed layer dilution using column water vapor (CWV, otherwise "precipitable water"). We noted that on multi-day timescales both water vapor and particles share near-surface sources and both fall to very low values with altitude; indeed, both are largely removed by precipitation. The existence of layers of H2O or aerosol not within the mixed layer adds complexity, but mixed-effects statistical regression captures essential proportionality of PM2.5 and the ratio variable (AOT ĝˆ• CWV). Accuracy is much higher than previous statistical models and can be extended to the whole Aqua satellite data record. The maps and time series we show suggest a repeated pattern for large valleys like the SJV - progressive stabilization of the mixing height after frontal passages: PM2.5 is somewhat more determined by day-by-day changes in mixing than it is by the progressive accumulation of pollutants (revealed as increasing AOT).