IWRA Proceedings

AbstractIn this work the effect of the time scale on the determination of natural flows in gauged hydrological basins is analyzed, using available hydrometric historic records and the mass conservation equation. The upper - middle basin of the Florido river (Rio Florido), in the northern Mexican State of Chihuahua, was used to perform this study. The basin was divided into the upper sub-basin (U-SB), with no infrastructure for water use, and the middle sub-basin (M-SB), where the well-infrastructured Irrigation District 103 is located. The mean annual natural flow for both sub-basins was estimated for the annual and monthly time scales, for the period spanning from 1982 to 2002. There are no return flows in U-SB, while in M-SB flows are generated by the irrigation district. By correcting the annual hydrogram of mean monthly natural flows, it was estimated that 30% of the irrigation water volume returned to the river during the analysis period, as is discussed in a parallel study (Silva-Hidalgo et al., 2008). The estimated natural flow was 116.8 Mm3 for the U-SB at the annual scale, and 119.9 Mm3 at the monthly scale. These results differ in only 2.65%, which shows that the time scale had very low impact. The mean annual natural flow in M-SB, calculated using a return flow of 30% as determined before, results in 59.87Mm3 at the monthly scale and of 49.6 Mm3 at the annual scale. These results are inconsistent, since they differ 20.7% from each other. This suggests that the time scale is indeed very important in basins where water is used for irrigation. The inconsistency of the results obtained at different time scales can be explained if it is observed that, during the calculations for both sub-basins, negative natural flows were obtained during several years for some of the dry months between February and June. A negative natural flow does not have physical meaning, but can be related to water losses along the river, or to water extractions larger than reported. The detection and correction of the anomaly, if it shows up, can only be achieved when a monthly or shorter time scale is used. The correction is performed by setting all negative calculated values to zero; the obtained results are more reliable than those obtained using the annual time scale. It is concluded that for basins with irrigation infrastructure, the monthly time scale should be used for natural flow determinations. The use of an annual scale is only advised for preliminary evaluations.