Скачать презентацию Surface Skin Temperature Soil Moisture and Turbulent Fluxes Скачать презентацию Surface Skin Temperature Soil Moisture and Turbulent Fluxes

65dbead1f277222c2041c093b5147ab4.ppt

  • Количество слайдов: 38

Surface Skin Temperature, Soil Moisture, and Turbulent Fluxes in Land Models Xubin Zeng, Mike Surface Skin Temperature, Soil Moisture, and Turbulent Fluxes in Land Models Xubin Zeng, Mike Barlage, Mark Decker, Jesse Miller, Cindy Wang, Jennifer Wang Dept of Atmospheric Sciences University of Arizona Tucson, AZ 85721, USA (a) A revised form of Richards equation (b) CLM 3 simulation versus MODIS skin T Consistent (c) Treatment of turbulence below and above canopy as well as snow burial of canopy (d) Vegetation and snow albedo data

Revised Richards Eq. Revised Richards Eq.

Decker and Zeng (2007) Decker and Zeng (2007)

CLM 3 offline tests over Sahara, southwest US and Tibet For July 1 -5, CLM 3 offline tests over Sahara, southwest US and Tibet For July 1 -5, 2003.

Cs = Cs, soil W + Cs, veg (1 – W) W = exp(– Cs = Cs, soil W + Cs, veg (1 – W) W = exp(– LAI) Zeng et al. (2005)

Dickinson et al. (2006) Dickinson et al. (2006)

Thought experiment: What would be the land zo and d If above-ground biomass disappears? Thought experiment: What would be the land zo and d If above-ground biomass disappears? CLM 3 deficiency: zo and d depend on vegetation type only Solution: de = d V + (1 – V) dg ln (zoc, e) = V ln(zoc) + (1 – V) ln (zog) V = (1 – exp[-β min(Lt, Lcr)])/(1 – exp[- β Lcr])

Impact in CLM 3 Impact in CLM 3

Figure C. 1 CLM 3 -simulated snow depth and surface fluxes from Jan. 11 Figure C. 1 CLM 3 -simulated snow depth and surface fluxes from Jan. 11 -13, 1996 over a boreal grassland site in Canada. Both simulation with new formulation of fv, sno and simulation with standard CLM 3 are shown (52. 16ºN, 106. 13ºW ).

Wang and Zeng (2007) Figure C. 2 The same simulation as in Fig. C. Wang and Zeng (2007) Figure C. 2 The same simulation as in Fig. C. 1 but for averaged diurnal cycles of winter time (Dec. 1995, Jan. and Feb. 1996).

Figure C. 4 (a) Ten-year averaged DJF differences of Tg between CLM 3 with Figure C. 4 (a) Ten-year averaged DJF differences of Tg between CLM 3 with Eq. (C. 3) and the standard CLM 3 global offline simulations, and (b) ten-year averaged annual cycle of Tg difference over Alaska (59 -72ºN, 170 -140ºW).

NCAR/CLM 3: FVC(x, y), LAI(x, y, t) NCEP/Noah: GVF(x, y, t), LAI=Const Validation: 1 NCAR/CLM 3: FVC(x, y), LAI(x, y, t) NCEP/Noah: GVF(x, y, t), LAI=Const Validation: 1 -3 m spy sat data, 1 -5 m aircraft data, 30 m Landsat data, Surface survey data Zeng et al. (2000)

Data Impact Barlage and Zeng (2004) Data Impact Barlage and Zeng (2004)

NLDAS GVF Data Noah 1/8 degree monthly Miller et al. (2006) MODIS 2 km NLDAS GVF Data Noah 1/8 degree monthly Miller et al. (2006) MODIS 2 km 16 -day

NLDAS GVF Results grass Addition of new GVF dataset results in an increase of NLDAS GVF Results grass Addition of new GVF dataset results in an increase of transpiration (up to 35 W/m 2) and canopy evaporation (up to 8 W/m 2) • Balanced by a decrease in ground evaporation (up to 20 W/m 2) • Overall increase in LHF(up to 20 W/m 2) is balanced by decreases in SHF(up to 10 W/m 2) and Lwup(5 W/m 2) • crop Miller et al. (2006)

MODIS versus Noah maximum snow albedo data Albedo NDSI NDVI Land cover Individual bands MODIS versus Noah maximum snow albedo data Albedo NDSI NDVI Land cover Individual bands Red: NN filled Blue: LAT filled Green: > 0. 84 Barlage et al. (2005)

Impact on NLDAS Offline Noah Tests Barlage et al. (2005) Impact on NLDAS Offline Noah Tests Barlage et al. (2005)

Application of MODIS Maximum Snow Albedo to WRF-NMM/NOAH • up to 0. 5 C Application of MODIS Maximum Snow Albedo to WRF-NMM/NOAH • up to 0. 5 C decreases in 2 -m Tair in regions of significant albedo change • > 0. 5 C increase in 2 -m Tair in several regions Barlage et al. (2007)

Summary • Skin temperature and turbulent fluxes are all strongly affected by the treatment Summary • Skin temperature and turbulent fluxes are all strongly affected by the treatment of below and above-canpy turbulence and snow burial • They are also affected by green vegetation cover data as well as maximum snow albedo data • While Terra/Aqua MODIS provides 4 skin Ts measurements a day, its use without constraint from Tair requires additional efforts • The revised Richards equation should be used for land models for improved simulations of soil moisture and fluxes

Suggestions on LANDFLUX • Try to reach some consensuses on the land boundary data Suggestions on LANDFLUX • Try to reach some consensuses on the land boundary data to be used • Identify flux tower sites with relatively comprehensive data over different climate regimes to set up minimum criteria for land models or model components to meet • Try to use land-atmosphere constrained land atmospheric forcing data

Model Run • Model Alterations – New Richards equation • Including new bottom boundary Model Run • Model Alterations – New Richards equation • Including new bottom boundary condition • NO TUNABLE PARAMETERS – Soil texture constant with depth – Infiltration – Area of Saturated Fraction • 1984 -2004 with Qian/Dai forcing

Comparison of CAM/CLM 3 with the Terra and Aqua MODIS data Comparison of CAM/CLM 3 with the Terra and Aqua MODIS data

Zeng et al. (2007) Zeng et al. (2007)

Fractional Vegetation Cover NCAR/CLM 3: FVC(x, y), LAI(x, y, t) NCEP/Noah: GVF(x, y, t), Fractional Vegetation Cover NCAR/CLM 3: FVC(x, y), LAI(x, y, t) NCEP/Noah: GVF(x, y, t), LAI=Const Validation: 1 -3 m spy sat data, 1 -5 m aircraft data, 30 m Landsat data, Surface survey data Histogram of evergreen Broadleaf tree NDVIveg = 0. 69

Interannual variability and decadal trend of global fractional vegetation cover from 1982 to 2000 Interannual variability and decadal trend of global fractional vegetation cover from 1982 to 2000 Zeng et al. (2003)

Maximum Snow Albedo in the NCEP Noah Land Model (a) Shading effect (b) Shadowing Maximum Snow Albedo in the NCEP Noah Land Model (a) Shading effect (b) Shadowing effect (c) LAI is difficult to measure in winter! (d) (e) A = Asn fsn + Av(1 -fsn) (f) Then the question is (g) (h) (i) (j) (1) what is satellite snow fraction? (2) What is Asn?

Issue: Consistency of Cx below/within canopy Motivation: warm bias of 10 K in Tg Issue: Consistency of Cx below/within canopy Motivation: warm bias of 10 K in Tg in CCSM 2 Below/within canopy in CLM Hg ~ C s u* (Tg – Tva) Hf ~ Cf LAI u*0. 5 (Tv – Tva) Cs = const in BATS, LSM, CLM 2 Based on K-theory Cs ≈ 0. 13 b exp(-0. 9 b)/[1 – exp(-2 b/3)] b = f(LAI, stability)

Surface Skin Temperature, Soil Moisture, and Turbulent Fluxes in Land Models Xubin Zeng Mike Surface Skin Temperature, Soil Moisture, and Turbulent Fluxes in Land Models Xubin Zeng Mike Barlage, Mark Decker, Jesse Miller, Cindy Wang, Jennifer Wang Dept of Atmospheric Sciences University of Arizona Tucson, AZ 85721, USA

Turbulence Energy Balance: Water Balance: Turbulent fluxes Rnet + G + Ft + Fq Turbulence Energy Balance: Water Balance: Turbulent fluxes Rnet + G + Ft + Fq ≈ 0 P ≈ Fq + R Fx ~ Cx U (Xa – Xs) Cx = f(Zom, Zot, stability) X: temperaure, humidity, wind, trace gas (a) Consistent treatment of turbulence below and above canopy as well as snow burial of canopy (b) Vegetation and snow albedo data (c) CAM 3/CLM 3 simulation versus MODIS skin T (d) A revised form of Richards equation