Paleohydrologic estimates of convective rainfall in the Rocky Mountains

Convective rainfall is characteristically localized and can have large gradients in both rain rates and rainfall amounts over very short distances, often of a few kilometers or less. In remote areas, systematic precipitation networks may be sparse or nonexistent. Thus limited opportunity exist to obtain rainfall bucket survey data, particularly in areas burned by wildfire. Paleoflood hydrology includes the study of flood-transported sediments and botanic information from past floods preserved in stream channels and is particularly useful in providing flood data from which hydrometeorologic information can be inferred for ungaged basins. Two paleohydrologic techniques to estimate rainfall amounts of convective rainstorms are being developed and evaluated. Increased emphasis on extreme rainstorm and flood documentation will maximize benefits from emerging technologies, will help improve hydrologic modeling capabilities, and improve flash-flood forecasting.

In the first or geomorphic method, rainfall amounts can be inferred from the amount of hillslope erosion, maximum size of sediments transported, and deposition characteristics, preferably on sparsely vegetated hillslopes. Before substantial additional rainstorms, mapping of the location and dimensions of fresh rills, gullies, and headcuts as well as maximum size of sediments transported and their deposition characteristics. Local residents can often provide valuable information about the rainstorm including bucket data and storm duration. Bucket data, where available in the study area, are used to calibrate the geomorphic data accounting for variability such as soil type and cohesiveness, vegetation cover, and hillslope gradient. Then, variations in the geomorphic characteristics in the area affected by the rainstorm are used to estimate relative rainfall amounts. Finally, an isohyetal map is drawn considering local topography. Sometimes, storm path can be determined.

In the second or hydrologic method, indirect estimates of peak discharge (flood) are obtained for many small basins in the area affected by the storm. High-water marks (HWMs) of recent floods or paleostage indicators (old HWMs) for paleofloods and hydraulic data for a stream site are used to estimate peak discharge. Rainfall-runoff modeling using hydrologic models and physical basin attributes are used to derive independent estimates of rainfall for each small basin. Hydrologic rainfall estimates can be used: (1) to help draw the isohyetal map with geomorphic rainfall estimates, if available; (2) to develop isohyetal maps for historic and prehistoric rainstorms; (3) as a comparison with other independent sources of rainfall data (gaged, bucket survey data if not used to calibrate the geomorphic method, and NEXRAD radar), or; (4) to provide rainfall information when no other source exists.

The paleohydrologic methodology is a flexible, storm-chasing approach that provides independent, cost-effective rainfall estimates. Results usually are obtained with less than two days combined field and office work (for a 100 to 150 km2 storm). Elements of this presentation include: (1) discussion of study approach, results, and benefits for recent, historic, and prehistoric rainstorm reconstructions in the Rocky Mountain region; (2) discussion of the effects of limitations and uncertainties of the paleohydrologic methodology; (3) discussion of transferring the study approach to other hydroclimatic regions, and; (5) identification of future research needs.