RGDSS_Task8-8_HistoricMonthlyModel_BaseflowCheckandSimulationRuns

Home Browse Search

As shown in the graphs of Attachment 3, modeled EOM contents are quite favorable in comparison to historic observations for most reservoirs and are very similar to results from Phase lb model runs. Seasonal trends (fill/release) are evident in both time-series and match well in most cases. In some instances, the model clearly reports releases from storage that apparently did not occur historically and vice-versa. EOM contents comparisons with model results at several reservoirs are less favorable, including Terrace, Smith and Mountain Home reservoirs. Differences between modeled and observed conditions at Terrace Reservoir are likely due to some of the same issues previously discussed regarding baseflows and simulated shortages on the Alamosa River -these situations result in additional modeled releases from Terrace. Differences between modeled and observed conditions on Trinchera Creek reservoirs (Mountain Home and Smith), suggest possible problems with baseflow calculations and determination of losses -again resulting in excessive modeled releases. Comparisons at Eastdale Reservoir, while poor early in the simulation period, actually agree quite favorably during the latest 10 years when actual EOM data are available (EOM was estimated during the earlier portion of the study period). EOM contents at all reservoirs will continue to be a focus of calibration efforts in later stages of model development and as additional data filling techniques are investigated. Compact Operations The Rio Grande Compact is operated in the Phase lc model according to two operating rules designed to mimic current operational practices by Division 3. Division 3 operations are based on forecasted inflows at several "index" gages, year-to-date actual flows, previous year's deliveries, and many years of experience and understanding of losses and gains along the river. Operating rules Type 17 and Type 18 have been implemented to ensure that delivery requirements defined by Article 14 of the Compact are met on the Rio Grande and Conejos River, respectively. The Compact operating rules are applied over the period 1969 to 1997 in the model simulation -this is the period over which current administration practices have been applied. It is important to note that because the "historic simulation" limits demands to their historic diversion amount operation of the Rio Grande compact tends to mimic historic operations that tried to pay back a debt to New Mexico beginning in the early 1970s. As shown in the figures provided as Attachment 2, modeled flows at the Rio Grande at Labatos are within 500 of (cl%) of historic flows. Additional details of the Compact simulation by the Surface Water Model are shown in Table 4. For each year from 1969 to 1997, this table shows, for both the Rio Grande and Conejos River, an accounting of flows at index gages, Compact delivery requirements, simulated deliveries, and whether there has been anover- orunder-delivery (surplus/shortage). The combined delivery requirement (obligation), total delivery to New Mexico and overall surplus/shortage status is also shown. The results shown in Table 4 indicate that the Compact operating rules implemented in the Phase lc Model are appropriately calling water past upstream diversion points to meet stateline delivery obligations. In any given year, there is typically an over- orunder-delivery of water; this is due to primarily to uncertainties about the locations and amounts of return flows that accrue to the rivers. Over the 29 years which the Compact rules operate in the model, there is a tendency for over-delivery because the "historic simulation" limits demands to historic diversions and Colorado was attempting to eliminate a significant debt accumulated prior to 1969 by curtailing C:Acdss\Task8-8.doc Phase lc Model September 27, 2000 -Page 13 of 17