Laserfiche WebLink
<br />24 hr SWE change ending 06Z, 17 Feb 2006 <br /> <br /> <br /> <br />, _;of2J.~i;? /' ~ <br />/ .... ;,<: "0_ <br />..,......... <br />'. - -woo. <br />...1. <br />',0 <br />/.,<ll.:l <br />-:T <br /> <br />. .' ~.;.1iO <br />dU"'~'''" <br /> <br />.. ~ <br /> <br />. <br />,..~ I~ <br />."['., ........] <br />.,!-o~ .; ,(~ ) <br />J: . <br />. ;i;:~ ~ <br />: ~9!: J-.> . ~~~\J <br />c _ ,~~. ;t.--UIt" ~~ <br />I __-<"-0: ~ ~ '~,' <br />. ~ \ <br />~ . <br />'-~~ -"'- tJ"::::'1:..'" <br /> <br />\ <br /> <br />Inc....of.._. <br />........... <br /> <br /> <br />.n <br />"..," <br />....." <br />........ <br />.... .. <br />..... .IIt <br />..... .. <br />.... 0" <br />.-.... <br />--...- <br />......... <br />....... <br />....... <br />...... <br />I""'" <br />........ <br />........ <br />u.. ., <br />." <br />c::J-.- <br /> <br />.--..- <br />-- <br /> <br />~ <br /> <br />""" <br />........... <br />........ <br />,..... <br />. , <br /> <br />Figure 5. An example of the dSWE r24 hr SWE change) for one day during a seeding event. The scale <br />at right shows dSWE color-coded in inches. The San Juan larget areas are shown by yellow polygons, <br />the lone Cone and Uncompahgre control areas by red and green polygons. respectively. These maps <br />were downloaded from the National Operational Hydrologic Remote Sensing Center website at <br />hltp:frwww.nohrsc.nws.govlinteractivefhtml/mao.html. <br /> <br />induced chllng\$ in dSWE, they were not evident in the dSWE average difTerence values (Fig. 3) <br />or any dSWE maps (Attachment U). <br /> <br />The existence of a seeding signal is not refuted by this analysis. although it is possible that it did <br />not exist. We are merely stating that such a signal was not evidenccd by this approach. Since <br />seeding is aimed at a target art'll and not just a poilU, we still believe that the use of a variable <br />that is averaged over target and control arellS (such as SNODAS) should be superior to a variable <br />measured at a point (as with SNOTEL). Nevertheless, the choice of a suitable variable ami <br />targetlc(lIltrol areas, however, is fraught with potential errors. That this analysis was unable to <br />disccrn any seeding signal might stem from this choice of target/control areas, errors in the <br />SNODAS model, data assimilation, forcing data, lack of model reso!utinn. or other factors. <br />Many prior studies have demonstrated the extreme ditliculty of discerning a small seeding signal <br />from much larger natural snow precipitation variability. It appears that this new approach, using <br />SNODAS. was also unable to overcome this ditliculty. Either SNODAS dSWE is "too blunt an <br />instrument" for this task, there was no seeding signal to be diagnosed, or it was masked by <br />storm-scale precipitation variability. <br /> <br />9 <br />