| Summary: | Previous work by Todd and Kniveton (2001) (TK2001) has indicated a statistically significant association (at the daily timescale) between short-term reductions in galactic cosmic rays, specifically Forbush decrease (FD) events, and reduced cloud cover, mainly over Antarctica (as recorded in International Satellite Cloud Climatology Project (ISCCP) D1 data). This study presents an extension of the previous work using an extended dataset of FD events and ISCCP cloud data over the period 1983-2000, to establish how stable the observed cloud anomalies are. Composite analysis of ISCCP data based on a sample of 32 FD events (excluding those coincident with solar proton events) indicates cloud anomalies with a very similar space/time structure to that previously reported, although of smaller magnitude. Substantial reductions in high level cloud (up to 12% for zonal mean, compared to 18% reported by TK2001) are observed over the high geomagnetic latitudes, especially of the southern hemisphere immediately following FD event onset. Largest anomalies are centred on the Antarctic plateau region during austral winter. However, the largest cloud anomalies occur where the accuracy of the ISCCP cloud retrievals is likely to be lowest, such that the results must be treated with extreme caution. Moreover, significant positive composite mean surface and tropospheric temperature anomalies centred over the same region are also observed for the FD sample from the National Center for Environmental Prediction (NCEP) reanalysis data. Such increased temperatures are inconsistent with the radiative effect of a reduction in high-level cloud during local winter. Overall, the results do not provide strong evidence of a direct galactic cosmic ray/cloud association at short timescales. The results highlight (a) the potential problems of data quality in the high latitude regions (b) the problems inherent in inferring cause and effect relationships from observational data alone (c) the need for further research to test competing hypotheses.
|
|---|