Click on “Download PDF” for the PDF version or on the title for the HTML version.

If you are not an ASABE member or if your employer has not arranged for access to the full-text, Click here for options.

Although the Runoff Curve Number method has been developed in the 1950s, and extensively used worldwide, it has not been so in Brazil, yet. The present study was conducted in a small experimental watershed in the southern central region of Cear State, Brazil (632S, 395W). The watershed is representative of caatinga, a type of scrubby forest, characteristic of the semiarid northeast in Brazil. The watershed has a mean annual rainfall of 800 mm, being 85% concentrated in the January to May period. All streams are ephemeral, and runoff persists until 24 hours after the causing rainfall event is over. Rainfall and streamflow records (64 rainfall-runoff events) were collected at a station located at the watershed outlet from 2008 to 2011. The representative watershed curve number, was determined by the following methods: a) arithmetic mean; b) geometric mean; c) median and c) asymptotic curve fit. For curve number computation, both unranked and ranked paired rainfall-runoff depth of each event was used. Results showed that the arithmetic mean value of curve number (78.8) is very close to the median value (79.4) and both are larger than the geometric mean (78.0). A smaller value of curve number (76.3), however, was obtained when using the asymptotic approach. The observed curve numbers, ranging from 73.0 to 86, decrease with increasing rainfall depth, and tend to a constant value towards larger rainfall events, which are the events that lead to flooding. The watershed presented a standard response.