Developing rating curves in the Ruvu river Sub-basin

The Florida International Research iWASH Initiative Limited commissioned this study to assess and develop rating curves in the Ruvu River Sub-basin. The work concentrated at the six key strategic stations within the Sub-basin that include: Ngerengere at Konga 1HA9A, Ngerengere at Mgude, Ruvu at Morogoro - Dar es Salaam Road 1H8, Ruvu at Kibungo 1H5, Mgeta at Mgeta 1HB2 and Mgeta at Duthumi 1HB1411. Specifically, the work reviews the existing rating curves for the existing stations on the Ruvu, update and validate cross-sections at the six strategic stations, conduct flow measurements, conduct analysis and produce new rating equations for the six strategic stations. The work utilizes the historical stage-discharge measurements and the new measurements collected during the period November 2012 to May 2013. Results indicate that most of the stations were established in the early 1950s and discharge measurement in these stations continued until late 1980s, with no measurements between early 1990s and mid 2000s. Review of the rating relationships indicates validity of historical data up to the early 1980s for most of the stations, but cannot be used to transform gauge heights (water levels) without modification. Assessment of physical characteristics at the six key stations revealed fair hydraulic controls. Nevertheless, special attention will be required to attend some observed threats at the channel hydraulic sections. Validation of the new developed rating curves at the key gauging stations indicate that stage- discharge relationships require updating by new current meter measurements, as for some stations the number of data points that were used to develop the rating relationships were too few. Data records reveal existence of outliers, which is likely attributable among others to use of un-calibrated current meter/sampling equipment, poor measurement techniques, inexperienced sampling technicians and/or computational errors. The new developed rating curves revealed absence of bias for all the stations except for 1H8. Of the six stations, only three stations namely 1H5, 1HB2 and 1HB1411 passed the ‘goodness of fit’ test, while the rest of the stations, 1HA15, 1HA9A and 1H8 the test criterion t was found to be consistently greater than the critical value, leading to the rejection of the null hypothesis of the assumption of random fluctuations. Therefore, the test detected the presence of abnormal long runs of positive or negative deviations. This most likely indicates presence of systematic trend in the deviations with time, indicating that the rating curves need adjustment for a shift in control. The study recommends for a need to increase the number of measured discharges at different stages so as to increase reliability of the rating curve equations. The findings are also considered to be indicative of the situation country wide, thus it is strongly recommend that the MoW (through its WSDP and other donors) make available additional funds to enable revisiting, testing and revision of rating curves on all major rivers to ensure higher confidence in the flow measurements.