Lab 2: Met Forcing at the Umpqua, Oregon Site

Transcription

1 Lab 2: Met Forcing at the Umpqua, Oregon Site

2 Exercise: Pascal Storck s Site

3 Part of the DEMO study where they cut dice patterns in the forest fun to find out plane window while flying to California. Pascal Storck s Ph.D. Thesis, 2000 Beneath canopy Clear-cut Shelterwood clearing Umpqua National Forest, NW of Crater Lake, Oregon

4 Tasks 1) Incoming LW for the clearing 1) Longwave sensors are expensive and break a lot, so we seldom have reliable observed data. First, ignore the observed LW. Choose two methods for esnmanng incoming LW irradiance from other met variables: plot them to compare, and then simulate snow in the Umpqua clearing. 1) OpNon A (advanced user): Generate your own forcing datasets using LW algorithms of your choice. 2) OpNon B (beginning user): Choose from the ready- to- go forcing datasets. 2) Plot the modeled snow against the observed snow. Do you think some esnmates are more correct than others? 3) Now use the observed LW compare with the esnmated LW and compare the model output using this forcing dataset. Does this confirm or change your earlier conclusions? 2) Met to use to model snow under the canopy (more on this later)

5 Clear sky longwave methods Much cited use Used in several Land Surface Models Used in Liston and Elder (2006) Used in Bohn et al. (2013) Recommended by Flerchinger et al.

6 Useful Resources for this assignment /se]ngs/umpqua/ EXCEL file: LW_methods_example_Dana.xlsx Matlab files: See deltat_fuse.m and rounnes it calls Forcing files (in /input/umpqua/): Storck_9697_Open_obsLW.txt Storck_9697_Open_estLW_DOKIA_cloudcorr.txt Storck_9697_Open_estLW_Idso.txt Storck_9697_Open_estLW_Prata.txt

7 Longwave at Umpqua, OR Cloud correction looks at ratio of obs to TOA SW Above ratio yields this Cut-offs in clearness index mark clouds

8 Special Issues with Forested Sites ExtrapolaNng data from a clearing to data above the forest top What issues do you think we should worry about?

9 Pascal Storck, 2000

10 Reference height = model forcing ~2 m height measure ments Pascal Storck, 2000

11 Reference height = model forcing Extrapolate wind to what you think it is above the canopy (assume logarithmic profile) ~2 m height measure ments Pascal Storck, 2000

12 Reference height = model forcing Then let the model extrapolate back down into the forest. Extrapolate wind to what you think it is above the canopy (assume logarithmic profile) ~2 m height measure ments Pascal Storck, 2000

13 Comparison of LW schemes at Umpqua

14 Longwave at Umpqua, OR Cloud correction looks at ratio of obs to TOA SW Above ratio yields this Clearly makes a big difference Cut-offs in clearness index mark clouds

15 Forest Winds Assignment Try simulanng under canopy snow with both adjusted above- canopy wind and non- adjusted above- canopy wind. Try adjusnng the model parameters for below- canopy wind ahenuanon

16 snow_zlocalahributes_! ***********************************************************************************! ***** DEFINITION OF SITE CHARACTERISTICS ******************************************! ***********************************************************************************! ***********************************************************************************! NOTES:!! 1) Lines starting with "!" are treated as comment lines -- there is no limit on th! 2) Ensure select from the **SAME TABLE** defined in the decisions file!! ***********************************************************************************! ***** NOTE: ensure select from the table used in the decisions file!! ***********************************************************************************! *********************************************************************************** hruindex HRUarea latitude longitude elevation tan_slope contourlength mheigh *******************************************************************************! ntourlength mheight vegtypeindex soiltypeindex slopetypeindex downhruindex! ! ! *******************************************************************************!! 7: Grassland! 13: Evergreen Broadleaf Forest

17 snow_zparamtrial_storcktree.txt! *********************************************************************************! ***** DEFINITION OF TRIAL MODEL PARAMETER VALUES ********************************! *********************************************************************************! *********************************************************************************! Note: Lines starting with! are treated as comment lines -- there is no limit on!!! Variable names are important: They must match the variables in the code, and they! NOTE: must include information for all HRUs!! ********************************************************************************* hruindex tempcritrain temprangetimestep heightcanopytop heightcanopybottom refinter ! ****************************************************! ottom refinterceptcapsnow snowunloadingcoeff ratiodrip2unloading windreductionparam! d ! d !

18 Wind ahenuanon through canopy = tunable parameter

19 Winds change below- canopy melt a lot