=Paper=
{{Paper
|id=Vol-2534/20_short_paper
|storemode=property
|title=Methods of Spatial Data Processing Based on Bayesian Approach for Environmental Monitoring
|pdfUrl=https://ceur-ws.org/Vol-2534/20_short_paper.pdf
|volume=Vol-2534
|authors=Ekaterina G. Klimova
}}
==Methods of Spatial Data Processing Based on Bayesian Approach for Environmental Monitoring==
https://ceur-ws.org/Vol-2534/20_short_paper.pdf
Methods of Spatial Data Processing Based on Bayesian Approach
for Environmental Monitoring
Ekaterina G. Klimova
Institute of Computational Technologies SB RAS, Novosibirsk, Russia, klimova@ict.nsc.ru
Abstract. One of the important tasks of monitoring of environment is the problem of ob-
taining the values of environmental parameters on a regular grid. At present, such prob-
lems are solved using all available observational data as well as the mathematical model
of the process of interest to us. The mathematical formulation of the problem is included
in the set of tasks of the so-called inverse modelling. If the probabilistic formulation of
the problem is considered, the Bayesian approach is applied. This approach is used in
popular algorithms, such as the ensemble Kalman filter, the ensemble Kalman smoothing,
the particle method. The report provides a brief overview of modern methods, as well as
approaches to their practical implementation.
Keywords: data assimilation, ensemble Kalman filter, satellite data
1. Introduction
One of the important problems of environment monitoring is the problem of obtaining values of environmental
parameters on a regular grid. At present, problems are solved using all available observational data as well as the
mathematical model of the process of interest to us. The mathematical formulation of the problem is included in the
set of problems of the so-called inverse modelling. The solution of the inverse modeling problem for a given process
model and a set of observational data cyclically in time is the task of data assimilation. The problem of inverse mod-
eling also includes the problem of estimation of the model parameters.
If the probabilistic formulation of the problem is considered, the Bayesian approach is applied. This approach is
based on popular algorithms, such as the ensemble Kalman filter, the ensemble Kalman smoothing, the particle meth-
od. If the considered random variables are Gaussian, and the forecast and observation models are linear, this problem
statement is equivalent to the variational statement of the data assimilation problem (4DVAR) [1].
The report provides a brief overview of modern data assimilation methods used in the environment modeling, as
well as approaches to their practical implementation.
2. The inverse modeling problem
The inverse modeling problem is the problem of estimation of the unknown vector of parameters x RM using
the vector of observational data y RP [2].
Let's suppose that
y M (x) δ ,
where M - the well-known prediction-observation operator, δ is a random noise with a given distribution function.
If the time series of observational data is considered, the change in time of the estimated variable is described using a
mathematical model. The task of data assimilation is usually understood as the time-sequential estimation of an un-
known quantity from observational data [2].
3. Bayesian approach to the data assimilation problem
Suppose that the time change of the estimated quantity xk is described by the model
xk 1 fk 1,k (xk ) ηk ,
k
where k is the time step number. In addition, observational data y are known:
y hk (x ) ε .
k k k
In these formulas ηk , ε k are random errors of forecast and observations, respectively, f k 1, k - model operator, hk -
transformation operator of the predicted variable into the observable one.
The Bayesian approach consists in applying the Bayes theorem to obtain the optimal estimate from observational
data and the prediction:
Copyright © 2019 for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 In-
ternational (CC BY 4.0).
p(y | x) p(x)
p(x | y) .
p(y)
Depending on which time period is needed for state estimation, it is possible to define three estimation problems:
p(xl | yk ,1 ), k l - forecast, p(xk | yk ,1 ) - filtration, p(xk ,0 | yk ,1 ) - smoothing, wheгe xk ,0 {xk , xk 1 , , x0 } ,
yk ,1 {yk , , y1} . Notation and definitions are taken from the review [1].
In the linear Gaussian case, the solution of the filtering problem is the Kalman filter, the solution of the smoothing
problem is the Kalman smoothing. In [3], it was proposed to use the Monte Carlo method for solving filtering and
smoothing problems, the so-called ensemble filtering and smoothing algorithms. In the ensemble Kalman filter in the
nonlinear case, the prediction of the covariance matrix is performed using a nonlinear model, and the condition of
Gaussianity is violated. Also in this case, the estimate at the analysis step is an approximation of the estimate of the
minimum variance. To solve the smoothing problem in the nonlinear case, iterative smoothing algorithms using en-
sembles are currently being considered [4]. As noted in [4], iterative methods provide a good approximation for
weakly non-linear models. In the nonlinear non-Gaussian case, the particle method is used, which is also based on the
Bayesian approach [1, 2].
4. The ensemble Kalman filter
The ensemble Kalman filter was proposed by Evensen G. [3]. Consider a nonlinear dynamic system as a process
equation
xtk f (xtk 1 ) ηtk 1
and an observation equation
yk h(xtk ) εtk ,
where h is the observation operator, generally speaking, non-linear, transforming the forecast values into the observa-
ble variable, ηtk 1 is the ‘model noise’ vector, ε tk is the vector of observation errors, xtk is the vector of estimated
variables T at the moment of time t k , ε tk and ηtk 1 are Gaussian random variables:
T
E[εk εk ] Rk , E[ηk 1 ηk 1 ] Qk 1 . We will assume xk to be a ‘true’ value.
t t t t t t t
The stochastic ensemble Kalman filter consists of an ensemble of forecasts {xkf ,n , n 1, , N}
xkf ,n f (xak ,n1 ) ηnk 1 (1)
and an ensemble of analysis {xk , n 1, , N}
a, n
xak ,n xkf ,n Kk (ynk εnk h(xkf ,n )) . (2)
The ensembles (1) and (2) provide a sample of ‘true’ values. Here the sample mean will be the optimal estimate,
and the deviations from the mean will be the ensembles of the analysis and forecast errors, respectively. To imple-
1 N f ,n errors {εk , n 1, , N} and
n
ment the ensemble version of the Kalman filter algorithm, an ensemble of observation
ensemble of forecast errors {dxk Txk xk , n 1, , N} , where xk xk , and an ensemble of model
f ,n f ,n f ,n f ,n
noise {ηnk 1 , n 1, , N} , E[ηnk 1 ηkn 1 ] Qk are specified. K k is a matrix ofNthen 1form
Kk Pkf HkT (Hk Pkf HkT Rk )1 ,
where Pkf and R k are matrices that are estimated by the ensemble averaging as
dxkf ,n dxkf ,n , Rk ε ε ,
1 N 1 N n n T
T
Pkf
N 1 n1 N 1 n1 k k
And H k is the linearized operator of h(xkf ,n ) with respect to xkf ,n :
h(xk ) h(xkf ,n ) Hk εkf .
Formulas (1) - (2) are a stochastic ensemble Kalman filter. The deterministic ensemble Kalman filter (analysis step)
consists of an equation for the mean
xak ,n xkf ,n Kk (y nk h(xkf ,n ))
and an estimate of the ensemble of analysis errors such that the corresponding covariance matrix satisfies the Kalman
filter equation Pka (I Kk Hk )Pkf [5].
5. The problem of optimal smoothing. Ensemble smoothing (EnKS - ensemble Kaman
smoother)
In the classical optimal estimation theory, the problem of optimal filtering is the problem of obtaining an optimal
estimate at the end of the considered time interval. Also, there is an optimal smoothing problem, which is the problem
of obtaining the optimal estimate for the given time interval. We suppouse, that an optimal estimate minimizes the
trace of the covariance matrix of estimation errors [6].
Ensemble algorithms are also applied for the optimal smoothing problem. In [3] was shown that the problem of
ensemble smoothing can be solved sequentially in time by using the available data at each time step to estimate val-
ues over the given time interval. In this case, the formulas for the optimal estimate are similar to those of the ensem-
ble Kalman filter. With this variant of ensemble smoothing (EnKS), calculations computing in the opposite direction
of time are not required. It should be noted that the EnKS algorithm is equivalent to the variational data assimilation
algorithm with the use of the corresponding weight matrices (4DVAR) [1, 2].
6. The parameters estimation in the data assimilation algorithm
Consider the equation of process as
xtk f (xtk 1 , αtk 1 ) ηtk 1 ,
observations are
yk h(xtk , αtk ) εtk ,
where αtk is the vector of parameters. We assume that the parameter does not change with time: αtk 1 αtk . Consider
the generalized estimation problem for vector z [x, α]T . Omitting the intermediate calculations, we write down the
result of the estimation procedure in general:
xa x f Pxx hxT (hx Pxx hxT R)1[y h(x f , α f )] Px hT (hx Pxx hxT R)1[y h(x f , α f )] ,
αa α f P x hxT (hx Pxx hxT R)1[y h(x f , α f )] P hT (hx Pxx hxT R)1[y h(x f , α f )] .
In these formulas, the index ‘k’ is omitted. The analysis step is considered, the index ‘a’ means analysis, the index ‘f’
means the prediction, P x - the cross-covariance of errors x and α , P - the covariance matrix of errors α , hx and
h - linearized operators with respect to x and α . If h does not depend on α , the estimation xa is carried out using
the same formula as in the classical Kalman filter [6]. In modern works on data assimilation, such an approach is ap-
plied to the estimation of greenhouse gas fluxes [7].
7. Variational approach to the data assimilation problem
The variational approach to the data assimilation problem is very popular and is used in the operational data
assimilation systems in the leading prognostic centers [1, 2]. The data analysis algorithm called 3DVAR (3-
Dimensional VARiational) consists in finding a vector x a that minimizes the functional
2J(x) (x xb )T B1 (x xb ) [y0 h(x)]T R1[y0 h(x)].
In this formula, B and R are the covariance matrices of forecast and observation errors, respectively. The operator h
is, generally speaking, nonlinear, transforms prognostic value to the observations (observed variables).
Minimum of functional is the solution of the equation
J(xa ) 0.
If the gradient of the functional is written in the following form:
J(x) B1 (x xb ) HT R1H(x xb ) HT R1{y0 h(xb )},
then the solution of the problem is
xa xb (B1 HTΡ1H){y0 h(xb )}.
Where H is the linearized operator h with respect to xb . This formula is equivalent to the analysis step of the Kalman
filter algorithm [1, 2, 6].
The algorithm 4DVAR is a generalization of 3DVAR to the space-time case. 4DVAR allows you to use obser-
vations from a certain time interval (t n t 0 ) . It is assumed that observations {yi0 ,i 0, , N} are specified at this
interval. It is required to find a minimum of functional
1 N
J[x(t 0 )] [x(t 0 ) xb (t 0 )]T B01[x(t 0 ) xb (t 0 )] [h(xi ) yi0 ]T R1[h( xi ) yi0 ]
1
2 2 i 0
provided that
x(t n ) Mn [x(t 0 )].
In this optimization problem, the control variable x(t 0 ) is the value at the initial time. That is, the algorithm 4DVAR
finds an initial condition such that a forecast with this initial value best approximates observations in a given time
interval. The minimum functional in 4DVAR is searched by iterative methods (for example, the quasi-Newton meth-
od). To implement the iterative process, we need to estimate the gradient of the functional J(x). The gradient of func-
tional is calculated using the conjugate linearized model. At present, there are data assimilation algorithms based on
4DVAR, with use the ensemble approach to estimate changes in covariances over time [1, 2].
8. Problems of implementation of ensemble algorithms
Tasks of data assimilation, the estimation of the parameters in the environment modelling has a very large dimen-
sion, requires huge computing resources. The use of ensemble algorithms partially solves this problem, but still the
problem remains extremely time-consuming. To carry out the analysis step of ensemble Kalman filter, algorithms of
transformation of the ensemble of forecasts are used to obtain the ensemble of the analysis. In this case, the analysis is
performed only for the ensemble mean value, and then the ensemble of analyses is calculated. There are deterministic
and stochastic variants of such algorithms for realization of both ensemble Kalman filter and ensemble Kalman
smoother [5].
A variant of a stochastic Kalman filter with ensemble transformations is the ensemble π-algorithm [8-10]. This al-
gorithm performs operations with matrices of the order of the ensemble dimension. In addition, this algorithm can be
used to implement ensemble smoothing (EnKS). An important property of the algorithm is its locality. The estimate
the desired value can be implemented independently for the given subdomains. This property can be used for estima-
tion of parameters locally in the given region.
When using ensembles in data assimilation, there are a number of problems associated with a small sample size
(ensemble size). In particular, it is a problem of high correlation values of forecast errors at long distances. In this
case, the so-called localization may be used, namely, the element-by-element multiplication of covariance matrices by
the function decreasing with the distance [1,5].
The small sample size is also one of the reasons for the divergence of the algorithm. The algorithms of the ensem-
ble Kalman filter (EnKF) and ensemble Kalman smoothing (EnKS) use the ‘inflation factor’. In this case the elements
of the covariance matrix are multiplied by a certain multiplier (multiplicative inflation), or an additional perturbation
is added to the ensemble of perturbations (additive inflation). It should be noted that these techniques are also used in
the iterative smoothing algorithms using ensembles, as well as in the particle method using ensembles [1, 2, 5].
Consider the features of the assimilation of satellite data [11]. Satellite data have a number of distinctive features
compared to ground-based and upper-air observations. These features include the following:
(a) The satellite measures radiation information, so it is necessary to convert model variables into observable vari-
ables using the radiation transfer equation. Thus, the observation operator h is nonlinear.
b) Observations are received continuously over time.
C) Observation errors correlate, the matrix of observational errors has large dimension and it is non-diagonal.
d) There is a of systematic error in observations (bias). This makes it difficult to use the formulas of the above al-
gorithms.
Satellite data can be used in two ways: directly, in the form of radiation data, or it is required to obtain meteoro-
logical values from satellite data in advance and then use them in the data assimilation procedure. The first option is
more preferable for a number of reasons:
(a) The transformation of radiation data uses forecast information, so the errors of the recovered data and fore-
cast will be correlated.
b) There is the problem of estimating the covariance matrix of observation errors.
Satellite data is a huge amount of information. However, as noted in a number of papers, a large number of ob-
servations do not always give the best result, since these data are an array of correlating random variables and
therefore they are not too informative [11].
9. The data assimilation in the environment modelling
The modern environmental monitoring is based on the mathematical modeling. In particular, models of the propa-
gation of greenhouse gases, active gas constituents, and aerosols in the atmosphere are considered.
Information on all these substances is measured using both ground-based measuring instruments and satellites. A
large number of investigations are currently devoted to the environmental data assimilation problem, using the basic
mathematical apparatus developed in the meteorological data assimilation [7].
An important task in monitoring of the environment is to assess the fluxes from the earth's surface of greenhouse
gases with the help of the data assimilation system [7].
The problems of the environment modeling are currently using models of the atmosphere with a "chemical"
block. As an example, the popular model WRF-Chem [12] may be considered. Modern databases of meteorological
fields (ERA-interim reanalysis [13], for example) are used for numerical experiments. ECMWF has created a MACC
database (Monitoring Atmospheric Composition and Climate), which includes active gas components, aerosols and
greenhouse gases [14]. In the regional modelling, an important problem is the determination of boundary values. Re-
analysis data may be used for this purpose. Also, in the literature, methods of data assimilation for determination val-
ues at the boundary of the region are considered [7].
10. Conclusion
The report provides a brief overview of current approaches to the problem of data assimilation in the environ-
ment modelling. The main aspects of practical realization of this task are considered.
References
[1] Carrassi A., Bocquet M., Bertino L., Evensen G. Data assimilation in the geosciences: An overview of meth-
ods, issuers and perspectives // Wiley interdisciplinary reviews: Climate Change. 2018. Vol. 131. Issue5, e535.
[2] Nakamura G., Potthast R. Inverse Modeling. IOP Publishing. 2015.
[3] Evensen, G. Data assimilation. The ensemble Kalman filter.Berlin Heideberg: Spriger-Verlag, 2009.
[4] Evensen G. Analysis of iterative ensemble smoother for solving inverse problems //Computational Geosci-
ences. 2018. Vol. 22. P.885-908.
[5] Houtekamer, H.L. Zhang, F. Review of the ensemble Kalman filter for atmospheric data assimilation //
Monthly Weather Review. 2016. Vol. 144. P. 4489-4532.
[6] Jazwinski, A.H. Stochastic processes and filtering theory. New York: Academic Press, 1970.
[7] Bocquet M.et al. Data assimilation in atmospheric chemistry models: current status and future prospects for
coupled chemistry meteorology models // Atmospheric Chemistry and Physics. 2015. Vol. 15. P. 5325-5358.
[8] Klimova, E. A suboptimal data assimilation algorithm based on the ensemble Kalman filter // Quarterly
Journal of the Royal Meteorological Society. 2012. Vol. 138. P. 2079-2085.
[9] Klimova E.G. Application of ensemble Kalman filter in environment data assimilation // IOP Conference Se-
ries: Earth and Environmental Science. 2018. Vol. 211. P. 012049.
[10] Klimova E G 2019 The Kalman stochastic ensemble filter with transformation of perturbation ensemble // Si-
berian. J. Num. Math. Vol. 22. No.1. P. 27-40.
[11] Thepaut J.-N. Satellite data assimilation in numerical weather prediction: an overview // Proceedings of the
ECMWF Seminar on Recent development in data assimilation for atmosphere and ocean, 8-12 September 2003.
2003. P.75-96.
[12] Grell G. et al. Fully coupled “online” chemistry within the WRF model // Atmos. Environment. 2005. Vol.
39. P. 6957- 6975.
[13] Dee D.P. et al. The ERA-interim reanalysis: Configuration and performance of the data assimilation system
// Q. J. Roy. Meteorol. Soc. 2011. Vol. 137. P. 553-579.
[14] Inness A. et al. The MACC reanalysis: an 8 yr data set of atmospheric composition // Atmos. Cgem. Phys.
2013. Vol. 13. P. 4073-4109.