Posted March 6, 2005, 7:35 AM.

Tom Chester of Fallbrook analyzed more than fifty years of rainfall data for Mt. Wilson, and found that a wetter February is highly correlated with a wetter March. And also vice versa, a drier February is correlated with a drier March. This doesn't absolutely true in every year, of course, but the correlation between the two months in any given year was high.

The last example against the trend, with a wet February and a dry March, occurred in 2001. In Temecula that year, 5.64" of rain fell in February and only 1.34" in March. Another notable exception was 1998.

Chester used a median sort to yield the results. This means he sorted the years by whether or not the February rainfall was above or below the normal amount. Then he averaged to see what above average February rainfall historically means for March, and what below average February rainfall historically means for March.

This scatterplot shows the correlation between the two months:

"The years with the February rainfall below the median averaged 2.13 inches of rain in March and the years with the February rainfall above the median averaged 8.56 inches of rain in March, " Chester said. "This shows dramatically the correlation."

So the statistics predicts that southern California should have a wet March this year, based on the wet February that we just had. These statistics are not alone in their assessment: The NOAA Climate Prediction Center is also predicting a wet March.