High-Resolution Image

The probability of rain often depends strongly on the hour of the day, especially over land. This image shows the favored hour for each of the 300-km boxes spanning the American continent and part of Africa, as well as nearby portions of the Atlantic and Pacific oceans. It is derived from six years (1998-2003) of TRMM data for the months of June, July, and August. The TRMM satellite is especially designed to detect rain, and it is in an orbit that causes it to visit areas at different hours of the day over the course of a month and a half.

The colors in the image are most intense when confidence in the maximum hour is high (see the colors along the top of the colorbar), and are darkened as confidence decreases. The gray areas are masked because they have low annual rainfall (lower than about 10 cm/year).

In the southeastern U.S., rain tends to fall most often in the afternoon, and later in the afternoon the more inland the area is. It tends to rain earlier in the day over the Gulf of Mexico. It is easy to understand why it might tend to rain most in the afternoon, since the land warms up during the course of a sunny day and rising hot air causes thunderstorms to develop easily by the afternoon. (There are a few spots deep within the U.S. where rain tends to occur closer to midnight, strangely enough - this is well documented.) It is interesting to compare this map with an analysis of day/night variations of rain over the U. S. from gauge data by the Climate Prediction Center at the National Centers for Environmental Prediction, NWS/NOAA.

Over much of the oceans rain events tend to happen before sunrise. The reasons can be complex. One possible reason is that clouds left over from the previous day cool after the sun has set, while the ocean remains warm, and it is the sinking of the cool cloud tops that triggers the rain events. Other processes involving cloud dynamics are probably also at work. Because many complex mechanisms are involved in the initiation of precipitation, reproducing maps like the one here presents an excellent challenge to models of the climate system. (Submitted by T. L. Bell and P. K. Kundu.)