IUKWC Workshop Nov16: Developing Hydro-climatic Services for Water Security – Session 5 – Item 3 J_Indu

LSM-RTM Coupling for
Microwave Tb Assimilation over
India
By
Dr. J. Indu
Assistant Professor
Department of Civil Engineering
I.I.T Bombay

Land Surface Models (LSM)
Indo-UK Workshop on 'Developing Hydro-Climatic Services for Water
Security', Nov 29 - Dec 1, 2016, Indian Institute of Tropical Meteorology
Pune, India
Configuration of the CLM subgrid hierarchy emphasizing the vegetation landunit
(Source: Adapted from NCAR Technical Note, 2010)
11/30/2016 2

The RTM is used as forward operator for assimilating the brightness temperature in LSM
The community Microwave Emission Model(CMEM)has been developed by European
Centre for Medium-Range Weather Forecasts (ECMWF) as the forward operator for low
frequency microwave brightness temperature from 1GHz to 20 GHz
TBtoa,p = TBau,p + exp(-τatm,p). TBtov,p
TBtov,p = TBsoil,p .exp (-τveg,p) + TBveg,p (1+ rr,p. exp(-τveg,p)) + TBad,p . rr,p .exp(-2.τveg,p)
Tbsoil,p = Teff .er,p
Tbveg,p = Tc . (1-ωp ). (1- exp(τveg,p ))
Community Microwave Emission Model (CMEM)
Pune, India
11/30/2016 3

Ensemble Kalman Filter
• Bayesian filtering process where an ensemble of model states are propagated
forward in time.
• The updated land surface states are also known as analyzed estimate which
is given as:
Where i is the grid number, j is the number of ensembles and U i,jt F i,jt
and O i jt are updated states, forecast states and observation state vectors
respectively, H is the observation operator (relating model states to
observation states and K is the Kalman gain matrix.
Pune, India
11/30/2016 4
, = , + ( , − , )
=
+

Data Assimilation Framework in LSM
• The propagation is made through the state equation
xi+1 = f t (xt , ut , wt)
• where, xt is the state vector of dimension n at time t, ft (.) is a dynamic
function, ut, and wt are the forcing variables and the process noise at time t,
respectively
• The state ensemble is propagated to the time step where a new measurement
becomes available.
• The measurement is related to the state via the measurement equation,
yt = ht (x t) + vt
• where, yt is the measurement vector of dimension m at time t, ht () is a
measurement transformation matrix and vt is independent measurement
noise. [Alemohammad, 2015]The conceptual diagram of the EnKF
(Source: Adapted from Alemohammad,2015)
Indo-UK Workshop on 'Developing Hydro-Climatic Services for
Water Security', Nov 29 - Dec 1, 2016, Indian Institute of
Tropical Meteorology Pune, India
11/30/2016 5

a). Whether satellite observed Tb circumvents the need for processing soil moisture
retrievals?
b). Relationship between the assimilation of Tb observations and the Radiative
Transfer Model (RTM)?
Motivation
Pune, India
11/30/2016 6

Data Used
Variable Spatial Resolution Temporal Resolution Source
Forcing Data
Near surface air temperature
0.47° × 0.47° 3 Hourly GDAS
Near surface specific humidity
Total incident shortwave radiation
Incident Longwave Radiation
Eastward wind
Northward wind
Surface pressure
Rainfall rate
Convective rainfall rate
LSM Parameters
Landcover 0.01 × 0.01 - AVHRR/UMD
Soil Texture 0.25 × 0.25 - FAO
Soil Fraction(clay,sand,silt) 0.25 × 0.25 - FAO
Slope type 0.01 × 0.01 - NCEP_LIS
Elevation SRTM
Albedo 0.01 × 0.01 Monthly NCEP_LIS
Greenness fraction 0.01 × 0.01 - NCEP_LIS
RTM Parameters
Soil fraction 0.25 × 0.25 - Ecoclimap/FAO
Geopotential 0.25 × 0.25 - NCEP
Vegetation fraction 0.25 × 0.25 - Ecoclimap
Vegetation Type 0.25 × 0.25 - Ecoclimap
[* LIS framework developed by Hydrological Sciences Laboratory at NASA’s Goddard Space Flight Center used for
assimilating the SMOPS soil moisture in Noah LSM v 3.6]
[* The Noah LSM for the present study is spun up by cycling seven times (2 years) through the period from 1 January 2008
to 31 December 2010 using the meteorological forcing from GDAS.]Indo-UK Workshop on 'Developing Hydro-Climatic Services for Water
Pune, India
11/30/2016 7

Taylor diagram for monthly mean surface soil moisture for each of JJAS months
from: (a,c,e,g) 2010; and (b,d,f,h) 2011
(+=Data Assimilated SSM; • = GLDAS SSM)
Cumulative Bias between DA and Openloop simulation: (a,c,e,g)
June,July,August,September (JJAS)2010; and (b,d,f,h) JJAS 2011.
[ Akhilesh and Indu (2016) ]
11/30/2016
Pune, India 8

Error Variance from TC: (a) DA; and (b) GLDAS
Pune, India
11/30/2016 9

Figure 1:TB simulation for ascending pass for parameterization 1: (a–d) AMSR-E H-polarized TB;
(e–h) simulated H-polarized TB; (i–l) AMSR-E V-polarized TB ; and (m–p) simulated V- polarized TB.
.
Figure 2.TB simulation for ascending pass for parameterization 2: (a–d) AMSR-E H- polarized TB;
(e–h) simulated H-polarized TB; (i–l) AMSR-E V-polarized TB; and (m–p) simulated V- polarized TB.
Pune, India
11/30/2016 10

Importance of Microwave Tb
 Direct assimilation of Tb observations in time-constrained
forecasting applications e.g. of hydrologic events, as it circumvents
the need for soil moisture retrieval data that are generally provided
with longer time-lag.
[ Lievens et al. (2015) ]
Pune, India
11/30/2016 11
[ Lievens et al. (2015) ]

Issues?
For assimilation of Tb directly into an LSM, the observations need to be unbiased
with respect to the model simulations (Reichle et al. 2004).
Strong seasonal cycle of Tb.
Complexity of Radiative Transfer processes involved [ De Lannoy et al. 2013).
Estimation of RTM Parameters?
Differing climatology between SM of LSM and satellite.
Pune, India
11/30/2016 12

Physics based model or statistical models?
Abramowitz et al. (2008) found that statistical models outperform
physics-based models at estimating land surface states and fluxes.
 Gong et al. (2013) provided a theoretical explanation for this result It
was shown that the extent to which the information available from
forcing data was unable to resolve the total uncertainty about the
predicted phenomena.
Pune, India
11/30/2016 13

Way Forward?
The amount of information contained in the forcing data = Total
amount of information actually made available to translate into
predictions/simulations [ Upper bound of Uncertainty ].
Our ability to resolve prediction problems will, to a large extent, be
dependent on our ability to collect and make use of observational data,
Model benchmarking scheme is required!!!
Pune, India
11/30/2016 14

THANK YOU FOR YOUR ATTENTION
Pune, India
11/30/2016 15
[Akhilesh S. Nair and J. Indu [2016], Enhancing Noah Land Surface Model Prediction Skill over Indian Subcontinent by
Assimilating SMOPS Blended Soil Moisture, Remote Sensing, 2016, 8, 976, doi:10.3390/rs8120976 (IF=3.03) ]

IUKWC Workshop Nov16: Developing Hydro-climatic Services for Water Security – Session 5 – Item 3 J_Indu

More Related Content

What's hot (20)

Similar to IUKWC Workshop Nov16: Developing Hydro-climatic Services for Water Security – Session 5 – Item 3 J_Indu (20)

More from India UK Water Centre (IUKWC) (20)

Recently uploaded (20)

IUKWC Workshop Nov16: Developing Hydro-climatic Services for Water Security – Session 5 – Item 3 J_Indu