Laserfiche WebLink
Memorandum To: Ray Alvarado Mch 10, 2003 <br />Page 2 <br />The first three sections address the daily vs. monthly modeling question more or less philos ~hically. <br />Section 4 of the memo is written for the user who intends to use StateMod and DMI switch $ to create <br />his own daily time step model from a monthly model, and faces decisions about how to reps, .sent <br />different parameters. Users who plan to use an existing daily model can profit from this sect pn as well. <br />because it may help them understand the advantages of, tradeoffs in, and rationale for the d 'qa <br />estimation techniques employed in the CDSS daily model. Section 5 summarizes the previo' s sections, <br />and Section 6 offers general observations and suggestions for future development of StateM d. <br />1.0 Deciding to use a daily time step model <br />StateMod now has the capability of operating on a daily time step, and daily time step Ovate resources <br />planning models are now available for use in several Colorado basins. The CDSS user is no faced with <br />deciding which type of model to appropriate for a given application. <br />The primary factor in the decision is generally the nature of the information sought from the model. <br />Both hydrologic phenomena and changes in the operating/demand environment very often ccur on a <br />sub-monthly basis, and are not fully characterized by a monthly model. Figures 1 and 2 sho <br />respectively how a daily time step model and a monthly time step model would allocate wat ,based on <br />the same basic data. The hydrograph in Figure 1 represents gaged daily streamflow for the River at <br />Clark, Colorado (USGS Station 09241000). Superimposed on the hydrograph are three hyp ~hetical <br />demands, which are "stacked" to reflect their seniority. The most senior is the yellow dem $, which <br />might be a winter instream flow requirement. The blue represents a more junior irrigation d mand <br />limited by decree to about 1080 cfs, and exhibits adry-up period for harvest during the seco d half of <br />August. The third, most junior demand, shown in green, is also for irrigation beginning Jun '1 s1 and <br />waning through July and August. Note that this relatively simple example excludes return fl s, and <br />therefore a demand can be met only when the hydrograph exceeds demand. Accordingly, th 'daily graph <br />(figure 1) indicates that the third most junior right would be able to divert for the first twely 'days of <br />June, as well as during late August, when it could take advantage of the middle right's dry- period. It <br />also shows that some of the peak that occurs in early June would not be diverted and would lbe available <br />to downstream users. <br />Figure 2 is based on the same underlying data, but daily values for both stream flow and de d have <br />been averaged for each month. According to the monthly interpretation, the most junior rig '. is never in <br />priority. Furthermore, the monthly model shows that the entire stream volume for the month bf June <br />could be diverted, while the daily model showed that some of the June peak would not be di 'erted and <br />would be available to downstream users. <br />The two figures illustrate that the daily model more accurately represents the yield of the ju for right, as <br />well as the state of the stream below the diversion during certain short periods. Therefore, if the issue <br />being analyzed specifically focuses on short-term conditions, like the yield of a flood right, en a daily <br />model may offer more accurate results than a monthly model. In general, modeling on a dail basis <br />could be preferable in investigations of these matters: <br />• General stream condition in "flashy" hydrologic systems <br />• Yield of new rights that capture only flood flows <br />CS-B99-190-0OB /TASKMEM03FiNAL.DOC ISO4~LE <br />