Hydrologic cycles Land Woods Hole, July 12-13, 2004

Transcription

1 ydrologic cycles and Woods ole, July -, 00 Alan etts, Atmospheric esearch, VT [EMWF] ettsydrologicycle00.pdf Are we losing the ability to make essential improvements in model physics because we are more concerned with fine tuning existing representations? [Easier if scientific supervision weak and resources diffused] Are we losing the ability (and perhaps the will) to make critical but often arduous tests of model physics against observations? [Need real-world link] Is the emphasis on quantifying model uncertainty diluting efforts to improve model physics? [Quantify uncertainty, when model interactions poorly understood!!]

2 The following questions: [ simulation versus understanding ]. What is the status of and what are the major errors associated with the parameterization of physical processes in atmosphere-ocean-land (A-O- ) models ranging from local-daily scales to regional-decadal scales? What effects do these errors have on model output compared to other sources of error? Global water cycle: the primary source of error is in the Tropics. oupling of WV,, E,, cloud-field, radiation field and processes.. Illustrate with Amazon. ow can model parameterizations be improved to represent the essential physics in A-O- models? ow can these parameterizations be tested rigorously, and what supporting infrastructure is needed to do so? ocally at points On river-basin scales oupling of observables/processes Where is the data? Was it assimilated [error field]? Do we have test data independent of model?. What is the appropriate balance between the efforts being directed toward improving physical parameterizations and the efforts being directed toward other model development and application activities? rocess: Quantify model errors on range of time and space scales Identify links/causes in either data assimilation or representation of physical processes asic research on new representations of physics - in parallel with pragmatic improvements Feed back to model development cycle; complete cycle in - years Appears simple, but very few centers actually complete this cycle [eg. EMWF] because it requires science-driven [as opposed to institutional] management and adequate resources/efficient utilization of resources

3 Simulation versus understanding?? False conflict: Earth system is too complex to understand without simulation. The real question is how well do we understand our simulations? [which requires a scientific attitude] Shift emphasis to: ow do the physical processes interact? ow does the coupled system behave in terms of observables?

4 0 W 0 W 0 W 0 W EWWF eanalysis EA-0 Amazon basins a) Amazon-mean b) : Madeira iver S 0 S 0 S 0 S 0 0 S 0 S 0 W 0 W 0 W 0 W

5 EA0 Annual means -00 : precipitation E: evaporation : runoff SM: olumn soil water, E, (mm day - ) 0 E E SM increment SM EA-0 Amazon Year

6 Annual means TWV: Total column water vapor : precipitation -bias from observations egression on TWV : precipitation : runoff -bias, ias +. (mm day - ), (mm day - ) EA-0 Amazon egression lines = -.+.*TWV = -.+.*TWV ias=0.*(twv-.) Year a : 0-h FX ias +. TWV analysis 0 TWV analysis (kg m - ) b : 0-h FX : 0-h FX : ias of 0 TWV analysis (kg m - )

7 EA0 Annual means corrected Mean is ± 0%, but ittle signal in interannual variability Data for uncertain, (mm day - ) 0 EA-0 Amazon Year : corrected (Marengo 00) (Dai et al.00) : corrected (Marengo 00)

8 Monthly precipitation and -bias from observations against TWV against TWV: [similar to annual] ias against data EA0 is wet in dry season dry in wet season [if -data is correct], ias from data (mm day - ) Amazon (monthly) :- :- :-00 Fit from Fig b ias from data TWV (kg m )

9 T ( o ) a Amazon Annual mean Dai et al. 00 EA0 T, T bias +. ( o ) b EA0 Amazon T T bias +. : 0-hFX : 0-h FX (mm day - ) Year Year. c Amazon Dai et al d Amazon EA0.0. T: data ( o ).. T: EA0 ( o ) T (T-.)=.*(-.) : data (mm day - ). T: EA0 (T-.)=0.*(-.) : 0-h FX (mm day - ) T control by precipitation

10 Surface-coupled physics 0 0 Madeira iver 0-00 a W net -0-0 (ha) EF Soil moisture index EF b Soil moisture index W net (Wm - ) Daily Means (Madeira iver) Soil moisture index against : of cloud-base EF: Evaporative fraction Soil moisture index against : ow cloud cover W net : surface net longwave

11 Tropospheric-coupled physics TWV (kg m - ) Madeira iver 0-00 TWV T Ω mid (ha day - ) a T (mm day - ) T 0.0*(TVWV-0.) TWV (kg m ) b T Daily means mid : mid-tropospheric omega against TWV and T TWV against recipitation and T [inear regression]

12 oupling of ascent with precipitation 0 E 0.*(Ω mid - 0) Madeira iver Madeira iver 0-00, E (mm day - ) 0 (mm day - ) 0 0 a Ω (ha day - ) mid 0 b : line TWV*(Ω -)/0 (mm day - ) mid Daily means and E against mid : mid-tropospheric omega Note <0 at mid 0ha/day Moist circulation and precipitation

13 Annual cycle- Madeira iver θ (K) 00 Madeira θ Q Q (g kg - ) (ha) TWV 0 0 Madeira TWV (kg m - ) Soil moisture index Soil moisture index loud cover T 0 loud over T Soil moisture index TWV (kg m - )

14 Annual cycle- SW net, net (W m - ) Madeira Soil moisture index SW net W net net W net (W m - ), E (mm day - ) Madeira TWV (kg m - ) E Madeira -00 W (ha) W net (W m - ) TWV (kg m - ) T T T T T T T T T T TWV T T E (mm day - ) T,

15 omparisons with data EA-0 point arvard forest tower

16 SW-cloud coupling to -Total cloud cover: EA0 -Transmitted fraction SW -Transmitted fraction A -compare W coupling