DENSE NETWORKS TO MEASURE CONVECTIVE RAINFALL IN THE SOUTHWESTERN UNITED-STATES

Almost all rangeland runoff in the southwestern United States results from intense short‐duration rainfall of limited‐area extent. In 1953, two southwest rangeland watersheds, the 150‐km2 Walnut Gulch and 174‐km2 Alamogordo Creek watersheds, were selected by U.S. Department of Agriculture Agricultural Research Service scientists as outdoor laboratories to measure water yield from rangeland watersheds. The networks were planned initially with gages on a 1‐mi grid, but difficulties in access resulted in more uneven networks. Because of budgetary restraints the basic networks were not completed until 1961. Gages were added after 1961, both for better estimates of rainfall on very small intensive study areas and to fill ‘gaps’ in the original network. The values of these dense rain gage networks became more apparent as the full complexity of southwestern rainfall became more apparent. Interstation correlations for storm rainfall decrease very rapidly with distance between stations. Inadequate networks can lead to significant errors in either underestimating or overestimating rainfall input as well as underestimating rainfall variability in verification of hydrologic models developed, for example, to estimate peak discharge, sediment production, or channel recharge. Finally, there are regional differences in intensities, areal extent, and duration of convective rainfall in the Southwest which require adequate sampling at several locations within the region. This paper is not subject to U.S. copyright. Published in 1979 by the American Geophysical Union.