Testing the applicability of neural networks as a gap-filling method using CH4 flux data from high latitude wetlands

Biogeosciences

Volumn 10, Issue 12, 2013, Pages 8185-8200

  Dengel, S ^a   Zona, D ^b,c   Sachs, T ^d   Aurela, M ^e   Jammet, M ^f   Parmentier, F J W ^g   Oechel, W ^c   Vesala, T ^a  

^a UNIVERSITY OF HELSINKI   (Finland)

^b UNIVERSITY OF SHEFFIELD   (United Kingdom)

^c SAN DIEGO STATE UNIVERSITY   (United States)

^d GFZ GERMAN RESEARCH CENTRE FOR GEOSCIENCES   (Germany)

^e FINNISH METEOROLOGICAL INSTITUTE   (Finland)

^f UNIVERSITY OF COPENHAGEN   (Denmark)

^g LUND UNIVERSITY   (Sweden)

Author keywords

[No Author keywords available]

Indexed keywords

ARTIFICIAL NEURAL NETWORK; EDDY COVARIANCE; GREENHOUSE GAS; LATITUDE; METHANE; SEASONAL VARIATION; TIME SERIES; WETLAND;

EID: 84890494740     PISSN: 17264170     EISSN: 17264189     Source Type: Journal    
DOI: 10.5194/bg-10-8185-2013     Document Type: Article

References (89)

1. Amari, S., Murata, N, Müller, K.-R., Finke, M., and Yang, M. H.: Asymptotic statistical theory of overtraining and crossvalidation, IEEE T. Neural Networ., 8, 985-996, 1997. (Pubitemid 127763330)
2. Anisimov, O. A.: Potential feedback of thawing permafrost to the global climate system through methane emission, Environ. Res. Lett., 2, 045016, doi:10.1088/1748-9326/2/4/045016, 2007.
3. Aubinet, M., Grelle, A., Ibrom, A., Rannik, U., Moncrieff, J., Foken, T., Kowalski, A. S., Martin, P. H., Berbigier, P., Bernhofer, C., Clement, R., Elbers, J., Granier, A., Grunwald, T., Morgenstern, K., Pilegaard, K., Rebmann, C., Snijders, W., Valentini, R., and Vesala, T.: Estimates of the annual net carbon and water exchange of forests: The EUROFLUX methodology, Adv. Ecol. Res., 30, 113-175, 2000.
4. Aurela, M., Lohila, A., Tuovinen, J. P., Hatakka, J., Riutta, T., and Laurila, T.: Carbon dioxide exchange on a northern boreal fen., Boreal Environ. Res., 14, 699-710, 2009.
5. Baldocchi, D. D.: Assessing the eddy covariance technique for evaluating carbon dioxide exchange rates of ecosystems, past, present and future, Glob. Change Biol., 9, 479-492, 2003. (Pubitemid 36540251)
6. Beck, L. W.: The Principle of Parsimony in Empirical Science, J. Phil., 40, 617-633, 1943.
7. Bubier, J. L., Moore, T. R., and Juggins, S.: Predicting Methane Emission from Bryophyte Distribution in Northern Canadian Peatlands, Ecology, 76, 677-693, 1995.
8. Bugmann, H. and Martin, P.: How Physics and Biology matter in forest gap models, Clim. Change, 29, 251-257, 1995.
9. Burnham, K. P. and Anderson, D. R.: Multimodel Inference, Understanding AIC and BIC in Model Selection, Sociological methods and research, 33, 261-304, 2004. (Pubitemid 39519124)
10. Carrara, A., Kowalski, A. S., Neirynck, J., Janssens, I. A., Yuste, J. C., and Ceulemans, R.: Net ecosystem CO2 exchange of mixed forest in Belgium over 5 years, Agr. Forest Meteorol., 119, 209- 227, 2003. (Pubitemid 37436087)
11. Chattopadhyay, S. and Chattopadhyay, G.: Identification of the best hidden layer size for three-layered neural net in predicting monsoon rainfall in India, J. Hydroinform., 10, 181-188, doi:10.2166/hydro.2008.017, 2008.
12. Christensen, T. R.: Methane emission from Arctic tundra, Biogeochemistry, 21, 117-139, 1993. (Pubitemid 24395621)
13. Cybenko, G.: Approximation by superpositions of a sigmoidal function, Math. Control Signal., 2, 303-314, 1989.
14. Dengel, S., Levy, P. E., Grace, J., Jones, S. K., and Skiba, U.: Methane emissions from sheep pasture, measured with an openpath eddy covariance system, Glob. Change Biol., 17, 3524- 3533, 2011.
15. Elizondo, D. and Góngora, M.: Current Trends on Knowledge Extraction and Neural Networks, Artificial Neural Networks: Formal Models and Their Applications - ICANN 2005, edited by: Duch, W., Kacprzyk, J., Oja, E., and Zadrozny, S., Springer Berlin, Heidelberg, 752-752, 2005.
16. EPA: Methane and Nitrous Oxide Emissions from Natural Sources, US, Environmental Protection Agency, Washington, DC, USA, 2010.
17. Eugster, W. and Plüss, P.: A fault-tolerant eddy covariance system for measuring CH4 fluxes, Agr. Forest Meteorol., 150, 841-851, 2010.
18. Falge, E., Baldocchi, D. D., Olson, R., Anthoni, P., Aubinet, M., Bernhofer, C., Burba, G., Ceulemans, R., Clement, R., Dolman, H., Granier, A., Gross, P., Grünwald, T., Hollinger, D., Jensen, N.-O., Katul, G., Keronen, P., Kowalski, A., Lai, C.-T., Law, B. E., Meyers, T., Moncrieff, J., Moors, E., Munger, J. W., Pilegaard, K., Rannik, Ü., Rebmann, C., Suyker, A., Tenhunen, J., Tu, K., Verma, S., Vesala, T., Wilson, K., and Wofsy, S.: Gap filling strategies for defensible annual sums of net ecosystem exchange, Agr. Forest Meteorol., 107, 43-69, 2001. (Pubitemid 32176659)
19. Forbrich, I., Gazovic, M., Kutzbach, L.,Wille, C.,Wolf, U., Becker, T., Schreiber, P., and Wilmking, M.: Cross-evaluation of measurements of peatland methane emissions on microform and ecosystem scales using high-resolution landcover classification and source weight modelling, Agr. Forest Meteorol., 151, 864- 874, 2011.
20. Franses, P. H. and Draisma, G.: Recognizing changing seasonal patterns using artificial neural networks, J. Econometrics, 81, 273- 280, 1997. (Pubitemid 127398243)
21. Fritsch, S. and Günther, F.: neuralnet, Training of Neural Networks, R package version 1.32, http://CRAN.R-project.org/package= neuralnet (05 April 2013), 2012.
22. Funk, D. W., Pullman, E. R., Peterson, K. M., Crill, P. M., and Billings, W. D.: Influence of water table on carbon dioxide, carbon monoxide, and methane fluxes from Taiga Bog microcosms, Global Biogeochem. Cy., 8, 271-278, 1994.
23. Gardner, M. W. and Dorling, S. R.: Artificial neural networks (the multilayer perceptron) - a review of applications in the atmospheric sciences, Atmos. Environ., 32, 2627-2636, 1998. (Pubitemid 28351824)
24. Gardner, M. W. and Dorling, S. R.: Neural network modelling and prediction of hourly NOx and NO2 concentrations in urban air in London, Atmos. Environ., 33, 709-719, 1999. (Pubitemid 29070976)
25. Gazovic, M., Kutzbach, L., Schreiber, P., Wille, C., and Wilmking, M.: Diurnal dynamics of CH4 from a boreal peatland during snowmelt, Tellus B, 62, 133-139, 2010.
26. Gorban, A., Rossiev, A., Makarenko, N., Kuandykov, N., and Dergachev, V.: Recovering data gaps through neural network methods, International Journal of Geomagnetism and Aeronomy, 3, 191-197, 2002.
27. Günther, F. and Fritsch, S.: neuralnet, Training of neural networks, R Journal, 2, 30-38, 2010.
28. Guan, L., Yang, J., and Bell, J. M.: Cross-correlations between weather variables in Australia, Build. Environ., 42, 1054-1070, 2007. (Pubitemid 44774656)
29. Gunaratnam, D. J., Degroff, T., and Gero, J. S.: Improving neural network models of physical systems through dimensional analysis, Appl. Soft Comput., 2, 283-296, 2003. (Pubitemid 40442504)
30. Hammerstrom, D.:Working with neural networks, Spectrum, IEEE, 30, 46-53, 1993. (Pubitemid 23711763)
31. Hansen, L. K. and Salamon, P.: Neural network ensembles, Pattern Analysis and Machine Intelligence, IEEE Transactions on, 12, 993-1001, 1990.
32. Hargreaves, K. J., Fowler, D., Pitcairn, C. E. R., and Aurela, M.: Annual methane emission from Finnish mires estimated from eddy covariance campaign measurements, Theor. Appl. Climatol., 70, 203-213, 2001. (Pubitemid 33098831)
33. Hecht-Nielsen, R.: Kolmogorov's mapping neural network existence theorem, P. 1st Int. Conf. Neural Network., 3, 11-14, 1987.
34. Hendriks, D. M. D., Dolman, A. J., van der Molen, M. K., and van Huissteden, J.: A compact and stable eddy covariance set-up for methane measurements using off-axis integrated cavity output spectroscopy, Atmos. Chem. Phys., 8, 431-443, doi:10.5194/acp-8-431-2008, 2008.
35. Hurvich, C. M. and Tsai, C. L.: Regression and Time Series Model Selection in Small Samples, Biometrika, 76, 297-307, 1989.
36. IPCC: Summary for Policymakers, in: Climate Change 2007, The Physical Science Basis, Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change, edited by: Solomon, S., Qin, D.,Manning, M., Chen, Z., Marquis, M., Averyt, K. B., Tignor, M. and Miller, H. L., Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA, 2007.
37. Jackowicz-Korczy'nski, M., Christensen, T. R., Bäckstrand, K., Crill, P., Friborg, T., Mastepanov, M., and Ström, L.: Annual cycle of methane emission from a subarctic peatland, J. Geophys. Res., 115, G02009, doi:10.1029/2008JG000913, 2010.
38. Jain, A. K., Mao, J., and Mohiuddin, K. M.: Artificial Neural Networks, A Tutorial. Computer, 29, 31-44, 1996.
39. Järvi, L., Nordbo, A., Junninen, H., Riikonen, A., Moilanen, J., Nikinmaa, E., and Vesala, T.: Seasonal and annual variation of carbon dioxide surface fluxes in Helsinki, Finland, in 2006-2010, Atmos. Chem. Phys., 12, 8475-8489, doi:10.5194/acp-12-8475- 2012, 2012.
40. Johansson, T., Malmer, N., Crill, P. M., Friborg, T., Åkerman, J. H., Mastepanov, M., and Christensen, T. R.: Decadal vegetation changes in a northern peatland, greenhouse gas fluxes and net radiative forcing. Glob. Change Biol., 12, 2352-2369, 2006. (Pubitemid 44857709)
41. Kettunen, A., Kaitala, V., Alm, J., Silvola, J., Nykänen, H., and Martikainen, P. J.: Cross-correlation analysis of the dynamics of methane emissions from a boreal peatland, Global Biogeochem. Cy., 10, 457-471, 1996.
42. Kettunen, A., Kaitala, V., Lehtinen, A., Lohila, A., Alm, A., Silvola, J., and Martikainen, P. J.: Methane production and oxidation potentials in relation to water table fluctuations in two boreal mires, Soil Biol. Biochem., 31, 1741-1749, 1999. (Pubitemid 29432552)
43. Kişi, O . and Oncuoǧlu, E. : Comparison of three back-propagation training algorithms for two case studies, Indian J. Eng. Mater. S., 12, 434-442, 2005. (Pubitemid 43156111)
44. Klevecka, I., and Lelis, J.: Pre-Processing of Input Data of Neural Networks: The Case of Forecasting Telecommunication Network Traffic, Telektronikk, 3, 168-178, 2008.
45. Lee, T. L. and Jeng, D. S.: Application of artificial neural networks in tide-forecasting, Ocean Eng., 29, 1003-1022, 2002. (Pubitemid 34276284)
46. Lek, S. and Guégan, J. F.: Artificial neural networks as a tool in ecological modelling, an introduction, Ecol. Model., 120, 65-73, 1999. (Pubitemid 29405129)
47. Long, K. D. and Flanagan, L. B.: Diurnal and seasonal variation in methane emissions in a northern Canadian peatland measured by eddy covariance, Glob. Change Biolo., 16, 2420-2435, 2010.
48. Mastepanov, M., Sigsgaard, C., Dlugokencky, E. J., Houweling, S., Ström, L., Tamstorf, M. P., and Christensen, T. R.: Large tundra methane burst during onset of freezing, Nature, 456, 628-630, 2008.
49. McDermitt, D., Burba, G., Xu, L., Anderson, T., Komissarov, A., Riensche, B., Schedlbauer, J., Starr, G., Zona, D., Oechel, W., Oberbauer, S., and Hastings, S.: A new low-power, open-path instrument for measuring methane flux by eddy covariance, Appl. Phys. B-Lasers O., 102, 1-15, 2011.
50. Moffat, A. M.: A new methodology to interpret high resolution measurements of net carbon fluxes between the terrestrial ecosystems and the atmosphere, Doctoral thesis, Friedrich Schiller University, Jena, 2010.
51. Moffat, A. M., Papale, D., Reichstein, M., Hollinger, D. Y, Richardson, A. D., Barre, A. G., Beckstein, C., Braswell, B. H., Churkina, G., Desai, A. R., Falge, E., Gove, J. H., Heimann, M., Hui, D., Jarvis, A. J., Kattge, J., Noormets, A., and Stauch, V. J.: Comprehensive comparison of gap-filling techniques for eddy covariance net carbon fluxes, Agr. Forest Meteorol., 147, 209- 232, 2007. (Pubitemid 47628782)
52. Moffat, A. M., Beckstein, C., Churkina, G., Mund, M., and Heimann, M.: Characterization of ecosystem responses to climatic controls using artificial neural networks. Glob. Change Biol., 16, 2737-2749, 2010.
53. Nelson, M., Hill, T., Remus, W., and O'Connor, M.: Time series forecasting using neural networks: should the data be deseasonalized first?, J. Forecasting, 18, 359-367, 1999.
54. Nguyen, H. and. Chan, C.: Multiple neural networks for a long term time series forecast, Neural Comput. Appl., 13, 90-98, 2004. (Pubitemid 38604234)
55. Ooba, M. and Hirano, T.: Comparisons of gap-filling methods for carbon flux dataset, A combination of a genetic algorithm and an artificial neural network, Ecol. Model., 198, 473-486, 2006. (Pubitemid 44416083)
56. Papale, D.: Data Gap Filling, edited by: Aubinet, M., Vesala, T., and Papale, D., Eddy Covariance, A Practical Guide to Measurement and Data Analysis, Springer Atmospheric Sciences, 159- 172, 2012.
57. Papale, D. and Valentini, R.: A new assessment of European forests carbon exchanges by eddy fluxes and artificial neural network spatialization, Glob. Change Biol., 9, 525-535, 2003. (Pubitemid 36536968)
58. Parmentier, F. J. W., van Huissteden, J., van der Molen, M. K., Schaepman-Strub, G., Karsanaev, S. A., Maximov, T. C., and Dolman, A. J.: Spatial and temporal dynamics in eddy covariance observations of methane fluxes at a tundra site in northeastern Siberia, J. Geophys. Res.-Biogeo., 116, G03016, doi:10.1029/2010JG001637, 2011a.
59. Parmentier, F. J. W., van der Molen, M. K., van Huissteden, J., Karsanaev, S. A., Kononov, A. V., Suzdalov, D. A., Maximov, T. C., and Dolman, A. J.: Longer growing seasons do not increase net carbon uptake in the northeastern Siberian tundra, J. Geophys. Res.-Biogeo., 116, G04013, doi:10.1029/2011JG001653, 2011b.
60. Peltola, O., Mammarella, I., Haapanala, S., Burba, G., and Vesala, T.: Field intercomparison of four methane gas analyzers suitable for eddy covariance flux measurements, Biogeosciences, 10, 3749-3765, doi:10.5194/bg-10-3749-2013, 2013.
61. R Core Team: R: A language and environment for statistical computing, R Foundation for Statistical Computing, Vienna, Austria, http://www.R-project.org/ (last access: 05.09.2013), 2013.
62. Riedmiller, M.: Rprop-description and implementation details, Technical Report, University of Karlsruhe, Germany, 1994.
63. Rinne, J., Riutta, T., Pihlatie, M., Aurela, M, Haapanala, S., Tuovinen, J.-P., Tuittila, E.-S., and Vesala, T.: Annual cycle of methane emission from a boreal fen measured by the eddy covariance technique, Tellus B, 59, 449-457, 2007. (Pubitemid 47193289)
64. Riutta, T., Laine, J., Tuittila, E.-S.: Spatial variation in plant community functions regulates carbon gas dynamics in a boreal fen ecosystem. Tellus B, 59, 838-852, 2007. (Pubitemid 47604520)
65. Roulet, N., Moore, T. I. M., Bubier, J., and Lafleur, P.: Northern fens, methane flux and climatic change, Tellus B, 44, 100-105, 1992.
66. Ryan, M., Müller, C., Di, H. J., and Cameron, K. C.: The use of artificial neural networks (ANNs) to simulate N2O emissions from a temperate grassland ecosystem, Ecol. Model., 175, 189-194, 2004. (Pubitemid 38708811)
67. Sachs, T., Giebels, M., Boike, J., and Kutzbach, L.: Environmental controls on CH4 emission from polygonal tundra on the microsite scale in the Lena river Delta, Siberia, J. Geophys. Res., 113, G00A03, doi:10.1029/ 2007JG000505, 2008.
68. Saxén, H. and Pettersson, F.: Method for the selection of inputs and structure of feedforward neural networks, Compu. Chem. Eng., 30, 1038-1045, 2006.
69. Schiffmann, W., Joost, M., and Werner, R.: Optimization of the backpropagation algorithm for training multilayer perceptrons, Technical report, University of Koblenz, Insitute of Physics, 1994.
70. Schmidt, A., Wrzesinsky, T., and Klemm, O.: Gap Filling and Quality Assessment of CO2 andWater Vapour Fluxes above an Urban Area with Radial Basis Function Neural Networks, Bound-Lay Meteorol., 126, 389-413, 2008.
71. Sharda, R. and Patil, R. B.: Connectionist approach to time series prediction: an empirical test, J. Intell. Manuf., 3, 317-323, 1992.
72. Stathakis, D.: How many hidden layers and nodes?, Int. J. Remote Sens., 30, 2133-2147, 2009.
73. Sturtevant, C. S., Oechel, W. C., Zona, D., Kim, Y., and Emerson, C. E.: Soil moisture control over autumn season methane flux, Arctic Coastal Plain of Alaska, Biogeosciences, 9, 1423-1440, doi:10.5194/bg-9-1423-2012, 2012.
74. Suyker, A. E., Verma, S. B., Clement, R. J., and Billesbach D. P.: Methane flux in a boreal fen: Season-long measurement by eddy correlation, J. Geophys. Res., 101, 28637-28647, doi:10.1029/96JD02751, 1996.
75. Svozil, D., Kvasnicka, V., and Pospichal, J.: Introduction to multilayer feed-forward neural networks, Chemometr. Intell. Lab., 39, 43-62, 1997. (Pubitemid 27512840)
76. Tagesson, T., Mölder, M., Mastepanov, M., Sigsgaard, C., and Tamstorf, M. P.: Land-atmosphere exchange of methane from soil thawing to soil freezing in a high-Arctic wet tundra ecosystem. Glob. Change Biol., 18, 1928-1940, 2012.
77. Takle, E. S, Massman, W. J., Brandle, J. R., Schmidt, R. A., Zhou, X., Litvina, I. V., Garcia, R., Doyle, G., and Rice, C. W.: Influence of high-frequency ambient pressure pumping on carbon dioxide efflux from soil. Agr. Forest Meteorol., 24, 193-206, 2004. (Pubitemid 38946014)
78. Toptygin, A. Y., Gribanov, K. G., Imasua, R., Bleuten, W., and Zakharovet, W. I.: Seasonal methane content in atmosphere of the permafrost boundary zone in Western Siberia determined from IMG/ADEOS and AIRS/AQUA data, Honolulu, HI, USA, SPIE, 2005.
79. Treadgold, N. K. and Gedeon, T. D.: The Sarprop Algorithm, A Simulated Annealing Enhancement To Resilient Back Propagation, Proceedings International Panel Conference on Soft and Intelligent Computing, 1996.
80. van Huissteden, J., Maximov, T. C., Kononov, A. V., and Dolman, A. J.: Summer soil CH4 emission and uptake in taiga forest near Yakutsk, Eastern Siberia, Agr. Forest Meteorol., 148, 2006- 2012, 2008.
81. van Wijk, M. T. and Bouten, W.: Water and carbon fluxes above European coniferous forests modelled with artificial neural networks, Ecol. Model., 120, 181-197, 1999. (Pubitemid 29405138)
82. Wang, W., van Gelder, P. H. A. J. M., and Vrijling, J. K.: Some Issues About the Generalization of Neural Networks for Time Series Prediction Artificial Neural Networks: Formal Models and Their Applications - ICANN 2005, edited by: Duch, W., Kacprzyk, J., Oja, E., and Zadrozny, S., Springer Berlin, Heidelberg, 749-749, 2005.
83. Whalen, S. C. and Reeburgh, W. S.: Interannual variations in tundra methane emission: A 4 year time series at fixed sites, Global Biogeochem. Cy., 6, 139-159, 1992.
84. Wille, C., Kutzbach, L., Sachs, T., Wagner, D., and Pfeiffer, E-M.: Methane emission from Siberian arctic polygonal tundra, eddy covariance measurements and modelling, Glob. Change Biol., 14, 1395-1408, 2008. (Pubitemid 351659559)
85. Windsor, J., Moore, T. R., and Roulet, N. T.: Episodic fluxes of methane from subarctic fens, Can. J. Soil Sci., 72, 441-452, 1993. (Pubitemid 23379788)
86. Zamolodchikov, D. G., Karelin, D. V., Ivaschenko, A. I., Oechel, W. C., and Hastings, S. J.: CO2 flux measurements in Russian Far East tundra using eddy covariance and closed chamber techniques, Tellus B, 55, 879-892, 2003. (Pubitemid 37196058)
87. Zhang, G. P. and Qi, M.: Neural network forecasting for seasonal and trend time series, Eur. J. Oper. Res., 160, 501-514, 2005.
88. Zona, D., Oechel, W. C., Kochendorfer, J., Paw, U. K. T., and Salyuket, A. U.: Methane fluxes during the initiation of a large-scale water table manipulation experiment in the Alaskan Arctic tundra, Global Biogeochem. Cy., 23, GB2013, doi:10.1029/2009GB003487, 2009.
89. Zona, D., Lipson, D. A., Paw, U K. T., Oberbauer, S. F., Olivas, P., Gioli, B., and Oechel, W. C.: Increased CO2 loss from vegetated drained lake tundra ecosystems due to flooding, Global Biogeochem. Cy., 26, GB2004, doi:10.1029/2011GB004037, 2012.