Article information

2014 , Volume 19, ¹ 4, p.69-83

Penenko A.V., Penenko V.V.

Direct data assimilation method for convection-diffusion models based on splitting scheme

Modern urban and industry areas emit large number of airborne chemical substances that can seriously pollute the environment. These processes can be studied with multidimensional convection-diffusion models. Both high spatial dimensionality and large variety of considered chemical species impose strong requirements to the performance of direct and inverse modeling algorithms. In this work, we consider a data assimilation problem for a nonstationary multidimensional convection-diffusion model with in situ concentration measurements. Data are assimilated at each time step of time discretized model. From theoretical point of view, data assimilation problem is considered as a sequence of inverse problems for different compositions of measurement data. Hence Tikhonov regularization can be a theoretical background to study the problem as an ill-posed measurement operator inversion problem with convection-diffusion model as a regularizer. The model is augmented with control functions that are added to the source term in order to resolve uncertainty of the emission rate inherent to the atmospheric chemistry models. They are sought as the minimum of the target functional in which a control function norm is combined with the misfit between measured and corresponding modeled values. Original element in the model lies in variational approach that has been applied to the weak-constraint formulation for each splitting stage of the multidimensional model. Because of the efficient algorithm for one-dimensional data assimilation to the convection-diffusion model, the resulting algorithm doesn't take iterations with given data assimilation parameter that regulates the measurements reproduction quality. This parameter can be chosen with Morozov discrepancy principle. Algorithm efficiency has been shown in computational experiments.

Aknowlegements: Our research is supported by RFBR through grants 14-01-31482-mol_a and 14-01-00125-a, the Presidium of RAS (Program 4), the Department of Mathematical Sciences of RAS (Program 3), the SB RAS integration projects 8 and 35.

Received 22 February 2014.

[full text]
Keywords: Data assimilation, variational approach, convection-diffusion, splitting method

Author(s):
Penenko Alexey Vladimirovich
PhD.
Position: The master of mathematics
Office: Novosibirsk State University, Institute of Computational Mathematics and Mathematical Geophysics SB RAS
Address: 6300090, Russia, Novosibirsk, Pirogova Str., 2
Phone Office: (383) 330-61-52
E-mail: a.penenko@yandex.ru
SPIN-code: 9950-8820

Penenko Vladimir Viktorovich
Dr. , Professor
Position: Head of Laboratory
Office: Institute of Computational Mathematics and Mathematical Geophysics of the Siberian Branch of RAS
Address: 630090, Russia, Novosibirsk, prospect Akademika Lavrentjeva, 6
Phone Office: (383) 330 61 52
E-mail: penenko@sscc.ru

References:
[1] Brajson A., Ju-Shi Ho. Prikladnaya teoriya optimal’nogo upravleniya. Moscow: Mir; 1972. 544. (In Russ.)
[2] Kalman R.E. A new approach to linear filtering and prediction problems. Trans. of the ASME-J. of Basic Eng. 1960; 82: ser. D. 35-45.
[3] Gordon N.J., Salmond D.J., Smith A.F.M. Novel approach to nonlinear/non-Gaussian Bayesian state estimation. IEE-Proceedings-F. 1993; 140: 107-113.
[4] Penenko V.V., Obraztsov N.N. Variacionnyj metod soglasovaniya polej meteorologicheskih jelementov. Meteorologiya i gidrologiya. 1976; 11: 1-11. (In Russ.)
[5] Penenko V.V. Metody chislennogo modelirovaniya atmosfernyh protsessov. Leningrad: Gidrometeoizdat; 1981. 352. (In Russ.)
[6] Dimet F.-X. L., Talagrand O. Variational algorithms for analysis and assimilation of meteorological observations: Theoretical aspects. Tellus. A. 1986; 38A: 97-110.
[7] Talagrand O., Courtier P. Variational assimilation of meteorological observations with the adjoint vorticity equation. Quart. J. Roy. Meteor. Soc. 1986; 113(I): Theory: 1311-1328.
[8] Elbern H., Strunk A., Schmidt H., Talagrand O. Emission rate and chemical state estimation by 4-dimensional variational inversion. Atmos. Chem. Phys. 2007; 7: 3749-3769.
[9] Anthes R.A. Data assimilation and initialization of hurricane prediction models. J. Atmos. Sci. 1974; 31: 702-719.
[10] Blum J., Le Dimet F.-X., Navon I.M. Data assimilation for geophysical fluids. Handbook of Numerical Analysis. Els. Sci. 2008; 14: 377-434.
[11] Navon I.M. Data assimilation for numerical weather prediction: A review. Data Assimilation for Atmospheric, Oceanic and Hydrologic Appl. 2009; 1: 21-65. (In Russ.)
[12] Data Assimilation: Making Sense of Observations. Eds. W. Lahoz, B. Khttatov, R. Menard. Heildelberg: Springer; 2010. 718 .
[13] Sandu A., Tianfeng C. Chemical data assimilation — an overview. Atmosphere. 2011; 2: 426-463.
[14] Freitag M.A., Potthast R.W.E. Synergy of inverse problems and data assimilation techniques. Large Scale Inverse Problems Comput. Methods and Appl. in the Earth Sci.: Radon Series on Comput. and Appl. Math. 2013; 13: 1-54.
[15] Penenko V.V. Variacionnye metody usvoeniya dannyh i obratnye zadachi dlya izucheniya atmosfery, okeana i okruzhajushhej sredy. Sib. zhurn. vychisl. matematiki. 2009; 12(4): 341-351. (In Russ.)
[16] Baklanov A., Schlunzen K., Suppan P., Baldasano J. et al. Online coupled regional meteorology chemistry models in Europe: Current status and prospects. Atmos. Chem. Phys. 2014; 14: 317-398.
[17] Agoshkov V.I. Teoriya i metody resheniya zadach variacionnoj assimilyacii obrazov. Moscow: IVM RAN; 2012. 147. (In Russ.)
[18] Tihonov A.N., Gonchlrskij A.V., Stepanov V.V., JagolA A.G. Chislennye metody resheniya nekorrektnyh zadach. Moscow: Nauka; 1990. 230. (In Russ.)
[19] Marchuk G.I. Metody vychislitel’noj matematiki. Moscow: Nauka; 1989. 608. (In Russ.)
[20] Samarskij A.A., Vabishhevich P.N. Additivnye shemy dlja zadach matematicheskoj fiziki. Moscow: Nauka; 2001. 306. (In Russ.)
[21] Bensusan A., Lions Zh.-L., Temam R. Metody dekompozicii, decentralizacii, koordinacii i ih prilozhenija. Metody vychislitel’noj matematiki. Novosibirsk: Nauka; 1975. 144-270. (In Russ.)
[22] Penenko V.V., Tsvetova E.A. Discrete-analytical methods for the implementation of variational principles in environmental applications . J. of Comput. and Appl. Math. 2009; 226: 319-330.
[23] Penenko A.V. Diskretno-analiticheskie shemy dlja resheniya obratnoj kojefficientnoj zadachi teploprovodnosti sloistyh sred gradientnymi metodami. Sib. zhurn. vychisl. matematiki. 2012; 15(4): 393-408. (In Russ.)
[24] Penenko A.V. Nekotorye teoreticheskie i prikladnye voprosy posledovatel’nogo variacionnogo usvoenija dannyh. Vychisl. tehnologii. 2006; 11. Specvypusk: Izbrannye dokl. Mezhdunar. konf. po izmereniyam, modelirovaniyu i informacionnym sistemam dlja izucheniya okruzhajushhej sredy. ENVIR0MIS-2006. Tomsk. 2006; 2: 35-40. (In Russ.)
[25] Marchuk G.I., Zalesny V.B. A numerical technique for geophysical data assimilation problems using Pontryagin’s principle and splitting-up method. Rus. J. of Numerical Analysis and Math. Modelling. 1993; 8(4): 311-326.
[26] Agoshkov V.I., Ipatova V.M., Zalesnyj V.B. i dr. Zadachi variacionnoj assimiljacii dannyh nabljudenij dlya modelej obshhej cirkulyacii okeana i metody ih resheniya. Izv. RAN. Fizika atmosfery i okeana. 2010; 46(6): 734-770.

Bibliography link:
Penenko A.V., Penenko V.V. Direct data assimilation method for convection-diffusion models based on splitting scheme // Computational technologies. 2014. V. 19. ¹ 4. P. 69-83
Home| Scope| Editorial Board| Content| Search| Subscription| Rules| Contacts
ISSN 1560-7534
© 2024 FRC ICT