Global climate model simulations show that substantial increases in water vapor accompany temperature increases and contribute a positive feedback to the basic temperature changes. The changes expected in moisture suggest that water vapor should be examined for evidence of increases. Radiosonde observations, despite shortcomings, provide the longest records of tropospheric water vapor. Increases in surface-500mb precipitable water during the period 1973-1993 using radiosonde data have been reported over North America (Ross and Elliott, 1996). Over the rest of the Northern Hemisphere since 1973, Elliott and Ross (1998) showed water vapor increases over China and the Pacific islands while the rest of Eurasia showed a mixture of positive and negative trends.
In this study we have extended the water vapor record using data beginning as early as 1948 from the CARDS project and examined changes over the longer period. However, the poor quality of early sensors at upper atmospheric levels and the resultant homogeneity problems when improved sensors were introduced reduce our confidence in total column estimates of precipitable water over the earlier period. To minimize these data quality and homogeneity issues that are most apparent in upper atmospheric layers, we have used lower level fields of dewpoint and specific humidity as surrogates for the total column water vapor.
The presentation will include discussion of the relationship between surface-500mb W and lower level humidity, the identification of homogeneous periods within the complete data record and water vapor trend estimates for the homogeneous periods.
References
Ross, R.J., and W.P. Elliott, 1996: Tropospheric water vapor climatology and trends over North America: 1973-93. J. Climate, 9, 3561-3574.
Elliott, W.P. and Rebecca J. Ross, 1998: Northern Hemisphere water vapor trends. 9th Symposium on Global Change Studies, Phoenix AZ, 39-41.