Statistics education in the atmospheric sciences

Analyses of atmospheric sciences data and models are heavily dependent upon statistical and probabilistic reasoning. Statistical methods have played a n important role in establishing physical relationships of atmosphere-ocean -land interactions and in the development and verification of numerical for ecast models. There is no reason to expect statistics to play a lesser role in the years ahead. Yet there has been and remains minimal formal training in statistics and probability for atmospheric sciences students in the Uni ted States. This is in sharp contrast to other fields such as the social sc iences. This paper describes issues surrounding the general weakness in for mal statistics education for both undergraduate and graduate atmospheric sc iences students, and the minimal collaboration between statisticians and at mospheric scientists. Recommendations are offered on statistics curriculum guidelines for undergraduate and graduate atmospheric sciences students, an d on improving the overall interaction between the statistics and atmospher ic sciences disciplines.