Advertisement

Singular Spectrum Analysis: Climatic Niche Identification

  • Irene L. Hudson
  • Marie R. Keatley
Chapter

Abstract

This chapter discusses singular spectrum analysis (SSA) and uses a 32 year record (1940–1971) of flowering of four eucalypt species (Eucalyptus leucoxylon, E. microcarpa, E. tricarpa and E. polyanthemos) to illustrate its use.

SSA delineated the trend, annual and biennial cycle in all four species. Additionally a 4 year cycle was detected in E. tricarpa. The trend and annual cycle were identified by SSA decomposition of the underlying climate profile (rainfall and mean, minimum, maximum temperatures).

An examination of the correlation between the reconstructed flowering series and lagged climatic components found that for E. leucoxylon and E. tricarpa there was a similar relationship to climate. These two species exhibit a significant negative relationship with the temperature variables and a positive relationship with rainfall. The strongest relationship for E. leucoxylon was with minimum temperature (ρ = –0.742). In E. tricarpa maximum temperature was the marginally stronger driver (ρ = –0.895). Both E. microcarpa and E. polyanthemos also share a similar relationship to climate but this differs to that of E. leucoxylon and E. tricarpa; in that these species were positively influenced at flowering by temperature and negatively by rainfall. For E. microcarpa minimum temperature is the main but weak influence (ρ = 0.383) and E. polyanthemos maximum temperature is the stronger influence (ρ = 0.674).

Keywords

Cyclicity Eigentriples Singular value decomposition Singular spectrum analysis Time series subcomponents 

Notes

Acknowledgments

We thank D Kupke, Hypatia Scholarship student in the School of Mathematics and Statistics, University of South Australia, for valuable editorial assistance in 2008 regarding the mathematical exposition. We also acknowledge the preliminary work of C Li in 2003, Honours student to Irene Hudson, at the University of Canterbury, Department of Mathematics and Statistics, Christchurch, New Zealand. Li’s Honours project “SSA – Theory and Application to Economic Time Series” inspired all our SSA ventures! Our work using SSA on eucalypts was started by K. Fukuda whose summer pilot project work in 2003 led to the joint presentation at the XXIInd International Biometric Conference: Hudson IL, Fukuda K, Keatley MR (2004) Detecting underlying time series structures and change points within a phenological dataset using SSA.

References

  1. Aban JLE, Tateishi R (2004) Application of Singular Spectrum Analysis (SSA) for the reconstruction of annual phenological profiles of NDVI time series data. Proceedings of Asian Association of Remote Sensing 2004 Section 11. Data Processing: data fusion. http://www.aars-acrs.org/acrs/proceedings2004.php (accessed 15th March 2008)
  2. Allen MR, Mutlow CT, Blumberg GMC et al. (1994) Global change detection. Nature 370:24-25CrossRefGoogle Scholar
  3. Ahas R, Aasa A, Menzel A et al. (2002) Changes in European spring phenology. Int J Climate 22:1727–1738CrossRefGoogle Scholar
  4. Allen MR, Smith LA (1996) Monte Carlo SSA: detecting irregular oscillations in the presence of color noise. J Climate 9:3373–3404CrossRefGoogle Scholar
  5. Allen MR, Smith LA (1997) Optimal filtering in singular spectrum analysis. Phys Lett 234:419–428CrossRefGoogle Scholar
  6. Ashton DH (1975) Studies of flowering behaviour in Eucalyptus regnans F. Muell Aust J Bot 23:399–411CrossRefGoogle Scholar
  7. Bassett OD (2002) Flowering and seed crop development in Eucalyptus sieberi l. Johnson and E. globoidea Blakely in a lowland sclerophyll forest in East Gippsland, Victoria. Aust For 65:237–254Google Scholar
  8. Beuhne FR (1914) The honey flora of Victoria. J Dept Agr Vic XII:610–618Google Scholar
  9. Blackman RB, Tukey JW (1958) The measurement of power spectra from the point of view of communication engineering. Dover, New YorkGoogle Scholar
  10. Broomhead DS, King GP (1986) Extracting qualitative dynamics from experimental data. Physica D 20:217–236CrossRefGoogle Scholar
  11. Colebrook JM (1978) Continuous plankton records – zooplankton and environment, northeast Atlantic and North sea, 1948–1975. Oceanol Acta 1:9–23Google Scholar
  12. Copland BJ, Whelan RJ (1989) Seasonal variation in flowering intensity and pollination limitation of fruit set in four co-occurring Banksia species. J Ecol 77:509–523CrossRefGoogle Scholar
  13. Danilov DL, Zhiglyavsky AA (1997) Principal components of time series: the caterpillar method. Saint Petersburg University Press, St Petersburg (in Russian)Google Scholar
  14. Davis GL (1969) Floral morphology and the development of the gametophytes in Eucalyptus stellulata Sieb. Aust J Bot 17:177–190CrossRefGoogle Scholar
  15. Dettinger MD, Ghil M, Strong CM et al. (1995) Software expedites singular-spectrum analysis of noisy time series. Eos Trans Amer Geophys Union 76:12Google Scholar
  16. D’Odorico PD, Yoo J, Jaeger S (2002) Changing seasons: an effect of the North Atlantic oscillation. J Climate 15:435–445CrossRefGoogle Scholar
  17. Doi H (2007) Winter flowering phenology of Japanese apricot Prunus mume reflects climate change across Japan. Climate Res 34:99–104CrossRefGoogle Scholar
  18. Dose V, Menzel A (2004) Bayesian analysis of climate change impacts in phenology. Global Change Biol 10:259-272CrossRefGoogle Scholar
  19. Dose V, Menzel A (2006) Bayesian correlation between temperature and blossom onset data. Global Change Biol 12:1451-1459CrossRefGoogle Scholar
  20. Elsner JB, Tsonis AA (1991) Do bidecadal oscillations exist in the global temperature record? Nature 353:551–553CrossRefGoogle Scholar
  21. Elsner JB, Tsonis AA (1996) Singular spectrum analysis: a new tool in time series analysis. Plenum Press, New YorkGoogle Scholar
  22. Fitter AH, Fitter RSR, Harris ITB et al. (1995) Relationships between first flowering date and temperature in the flora of a locality of central England. Func Ecol 9:55–60CrossRefGoogle Scholar
  23. Flint AW, Fagg PC (2007) Mountain ash in Victoria’s state forests, silviculture reference manual no. 1. Department of Sustainability and Environment, MelbourneGoogle Scholar
  24. Fraedrich K (1986) Estimating the dimension of weather and climate attractors. J Atmos Sci 43:419–432CrossRefGoogle Scholar
  25. Fukuda K (2004) New improved methods for application and interpretation of SSA: a case study of climate and air pollution in Christchurch, New Zealand, Dissertation, University of Can-terbury, ChristchurchGoogle Scholar
  26. Fukuda K, Hudson IL (2005a) Global and local climatic factors on sulfur dioxide levels: comparison of residential and industrial sites. In: Francis AR, Matawie KM, Oshlack A et al. (eds) Statistical Solutions to Modern Problems Proceedings of the 20th International Workshop on Statistical Modelling Sydney, AustraliaGoogle Scholar
  27. Fukuda K, Hudson IL (2005b) Investigations of short-term (hourly) weather influences on CO, NO, NO2, PM10 and SO2 Levels in Christchurch, New Zealand. In: Proceedings of the International Conference on Research Highlights and Vanguard Technology on Environmental Engineering in Agricultural Systems, Ishikawa, Japan, 12–15 SeptemberGoogle Scholar
  28. Fukuda K, Hudson IL, Pearson K (2004) Singular Spectrum Analysis combined with an Enhanced Fourier expansion (EFE) method: a case study of the impact of notable global and local weather events on air pollution in Christchurch, NZ. Paper presented at the American Statistical Association Computational Environmetrics Conference, Chicago, 21–23 OctoberGoogle Scholar
  29. Ghil M, Vautard R (1991) Interdecadal oscillations and the warming trend in global temperature time series. Nature 350:324-327CrossRefGoogle Scholar
  30. Ghil M., Yiou P (1996) Spectral methods: What they can and cannot do for climatic time series. In Anderson D, Willebrand J (eds.) Decadal climate variability: dynamics and predictability. Elsevier, Amsterdam, pp 445-482Google Scholar
  31. Ghil M, Taricco C (1997) Advanced spectral analysis methods. In: Castagnoli GC, Provenzale A (eds) Past and present variability of the solar-terrestrial system: measurement, data analysis and theoretical models. Societa Italiana di Fisica, Bologna and IOS Press, AmsterdamGoogle Scholar
  32. Ghil M, Allen MR, Dettinger MD et al. (2002) Advanced spectral methods for climate time series. Rev Geophys 40:1–41CrossRefGoogle Scholar
  33. Golyandina N, Nekrutkin V, Zhigljavsky A (2001) Analysis of time series structure: SSA and related techniques. Chapman and Hall/CRC, Boca RatonCrossRefGoogle Scholar
  34. Golyandina N, Osipov E (2007) The “Caterpillar” – SSA method for analysis of time series with missing values. J Stat Plan Infer 137:2642–2653CrossRefGoogle Scholar
  35. Goodman RD (1973) Honey flora of Victoria. Department of Agriculture, MelbourneGoogle Scholar
  36. Gordo O, Sanz JJ (2005) Phenology and climate change: a long-term study in a Mediterranean locality. Oecologia 146: 484–495CrossRefPubMedGoogle Scholar
  37. Grigorov MG (2006) Global dynamics of biological systems from time-resolved omics experiments. Bioinformatics 22:1424–1430CrossRefPubMedGoogle Scholar
  38. Hassani H (2007) Singular spectrum analysis: methodology and comparison. J Data Sci 5:239–257Google Scholar
  39. Hannachi A, Jolliffe IT, Stehenson DB et al. (2005) In search of simple structures in climate: simplifying EOFs. Int J Climatol 26:7–28CrossRefGoogle Scholar
  40. Hudson IL, Fukuda K, Keatley MR (2004) Detecting underlying time series structures and change points within a phenological dataset using SSA. In: XXIInd International Biometric Conference Cairns, AustraliaGoogle Scholar
  41. Hudson IL, Keatley MR, Roberts AMI (2005a) Statistical methods in phenological research. In: Francis AR, Matawie KM, Oshlack A et al. (eds) Statistical solutions to modern problems. Proceedings of the 20th International Workshop on Statistical Modelling Sydney, AustraliaGoogle Scholar
  42. Hudson IL, Fukuda K, Dalrymple M (2005b) Climate-pollution impacts on Sudden Infant Deaths (SIDS): via SSA. In: Zerger A, Argent RM (eds) MODSIM 2005 International Congress on Modelling and Simulation. Modelling and Simulation Society of Australia and New ZealandGoogle Scholar
  43. Hudson IL, Kim SW, Keatley MR (2009) Climatic influences on the flowering phenology of four eucalypts: a GAMLSS approach. In: Anderssen RS, Braddock RD, Newham LTH (eds) 18th IMACS World Congress – MODSIM09 International Congress on Modelling and Simulation, Cairns, AustraliaGoogle Scholar
  44. Hsieh WW, Tang B (1998) Applying neural network models to prediction and data analysis in meteorology and oceanography. B Am Meterol Soc 79:1855–1870CrossRefGoogle Scholar
  45. Hsieh WW, Wu A (2001) Nonlinear multichannel singular spectrum analysis of the tropical pacific climate variability using a neural network approach. J Geophys Res-Oceans 107:13.11–13.15Google Scholar
  46. Hsieh WW, Wu A (2002) Nonlinear singular spectrum analysis. Neural Networks 3:2819–2824Google Scholar
  47. Jollife IT (1986) Principal component analysis. Springer, Berlin Heidelberg New YorkGoogle Scholar
  48. Keatley MR, Hudson IL (2000) Influences on the flowering phenology of three eucalypts. In: de Dear RJ, Kalma JD, Oke TR et al. (eds) Biometeorology and urban climatology at the turn of the century selected papers from the conference ICB-ICUC’99. World Meteorological Organisation, Geneva, SwitzerlandGoogle Scholar
  49. Keatley MR, Fletcher TD, Hudson IL et al. (2002) Phenological studies in Australia: potential application in historical and future climate analysis. Int J Climate 22:1769-1780CrossRefGoogle Scholar
  50. Keatley MR, Hudson IL, Fletcher TD (2004) Long-term flowering synchrony of box-ironbark eucalypts. Aust J Bot 52:47-54CrossRefGoogle Scholar
  51. Keatley MR, Hudson IL (2007) A comparison of the long-term flowering patterns of Box-ironbark species in Havelock and Rushworth forests. Environ Model Assess 12:279–292CrossRefGoogle Scholar
  52. Kim SW, Hudson IL, Keatley MR (2005) Mixture transition distribution analysis of flowering and climatic states. In: Francis AR, Matawie KM, Oshlack A et al. (eds) Statistical Solutions to Modern Problems Proceedings of the 20th International Workshop on Statistical Modelling Sydney, AustraliaGoogle Scholar
  53. Kondrashov D, Ghil M (2006) Spatio-temporal filling of missing points in geophysical data sets. Nonlin Proc Geophys 13:151–159CrossRefGoogle Scholar
  54. Kondrashov D, Ghil M (2007) Reply to T Schneider’s comment on “spatio-temporal filling of missing points in geophysical data sets”. Nonlin Proc Geophys 14:3–4CrossRefGoogle Scholar
  55. Kondrashov D, Feliks Y, Ghil M (2005) Oscillatory climate modes in extended Nile river record (AD 622–1922). Geophys Res Lett 32:L10702 doi:10.1029/2004GL022156CrossRefGoogle Scholar
  56. Kumaresan R, Tufts DW (1980) Data-adaptive principal component signal processing. In: Proc Conf on decision and control IEEE, Albuquerpque, pp 949–954Google Scholar
  57. Loeuille N, Ghil M (2004) Intrinsic and climatic factors in North-American animal population dynamics. BMC Ecol 4:6 doi:10.1186/1472–6785–4–6CrossRefPubMedGoogle Scholar
  58. Mac Nally R, Bennett AF, Thomson JR et al. (2009) Collapse of an avifauna: climate change appears to exacerbate habitat loss and degradation. Diversity Distrib 15:1–11CrossRefGoogle Scholar
  59. Mann ME, Lees JM (1996) Robust estimation of background noise and signal detection in climatic time series. Clim Change 33:409–445CrossRefGoogle Scholar
  60. Menzel A (2003) Plant phenology “Fingerprints”. Tasks for vegetation science, vol 39. Kluwer Academic Publishers, The NetherlandsGoogle Scholar
  61. Mineva A, Popivanov D (1966) Method of single trial readiness potential identification, based on singular spectrum analysis. J Methods 68:91–99Google Scholar
  62. Muir AM, Edwards SA, Dickins MJ (1995) Description and conservation status of the vegetation of the Box-ironbark ecosystem in Victoria. Department of Conservation and Natural Resources, MelbourneGoogle Scholar
  63. Paton DC, Crossfield EL, Hurrell B et al. (2004) Floral resources used by the South Australian apiary industry. Rural Industries Research and Development Corporation, Barton, ACTGoogle Scholar
  64. Paloma MJ, Sanchis R, Verdu G et al. (2003) Analysis of pressure signals using a singular system analysis (SSA) metholodology. Prog Nucl Energ 43:329–336CrossRefGoogle Scholar
  65. Paluš M, Novotná D (2004) Enhanced Monte Carlo Singular System Analysis and detection of period 7.8 years oscillatory modes in the monthly NAO index and temperature records. Nonlin Proc Geophy 11:721–729CrossRefGoogle Scholar
  66. Pike ER, McWhirter JG, Bertero M et al. (1984) Generalized information theory for inverse problems in signal processing. IEEE Proc 131:660–667Google Scholar
  67. Porter JW (1978) Relationships between flowering and honey production of Red ironbark, Eucalyptus sideroxylon (A. Cunn.) benth., and climate in the Bendigo district of Victoria. Aust J Agric Res 29:815–829CrossRefGoogle Scholar
  68. Roberts AMI (2008) Exploring relationships between phenological and weather data using smoothing. Int J Biometeorol 52:463–470CrossRefPubMedGoogle Scholar
  69. Rodó X, Pascual M, Fuchs G et al. (2002) ENSO and cholera: a nonstationary link related to climate change. PNAS 99:12901–12906CrossRefPubMedGoogle Scholar
  70. Rozynski G, Larson M, Pruszak Z (2001) Forced and self-organized shoreline response for a beach in the Southern Baltic sea determined through singular spectrum analysis. Coast Eng 43: 41–58CrossRefGoogle Scholar
  71. Salmerón M, Ortega J, García C et al. (2002) SSA, SVD, QR-cp, and RBF model reduction Lect Notes Comput Sci 2415:589–594CrossRefGoogle Scholar
  72. Schneider T (2001) Analysis of incomplete climate data: estimation of mean values and covariance matrices and imputation of missing values. J Climate 14:853–871CrossRefGoogle Scholar
  73. Schneider T (2007) Comment on “spatio-temporal filling of missing points in geophysical data sets” by D Kondrashov and M Ghil, Nonlin Processes Geophys 13, 151–159, 2006. Nonlin Proc Geophy 14:1–2CrossRefGoogle Scholar
  74. Schollhamer D (2001) Singular spectrum analysis for time series with missing data. Geophys Res Lett 16:3187–3190CrossRefGoogle Scholar
  75. Setterfield SA, Williams RJ (1996) Patterns of flowering and seed production in Eucalyptus miniata and E. tetradonta in a tropical savanna woodland, Northern Australia. Aust J Bot 44:107–122CrossRefGoogle Scholar
  76. Shun T, Duffy C (1999) Low-frequency oscillations in precipitation, temperature, and run-off on a west facing mountain front: a hydrologic interpretation. Water Resour Res 35:191–201CrossRefGoogle Scholar
  77. Schwartz MD, Reiter BE (2000) Changes in North American spring. Int J Climatol 20:929–932CrossRefGoogle Scholar
  78. Schwartz MD, Reed BC, White MA (2002) Assessing satellite derived start-of-season (SOS) measures in the conterminous USA. Int J Climatol 22:1793–1805CrossRefGoogle Scholar
  79. Schwartz MD (2003) Introduction. Phenology: an integrative environmental science. Tasks for vegetation science, vol 39. Kluwer Academic Publishers, The NetherlandsGoogle Scholar
  80. Slepian D (1978) Prolate spheroidal wave-functions, Fourier-analysis and uncertainty. 5. Discrete case. Bell Syst Tech J 57:1371–1430Google Scholar
  81. Somerville D, Campbell S (1997) Beekeeping in the Narrandera State Forests. N.S.W Agriculture, Goulburn, AustraliaGoogle Scholar
  82. Stöckli R, Rutishauser T, Dragoni D et al. (2008) Remote sensing data assimilation for a prognostic phenology model. J Geophys Res 113:G04021 doi:10.1029/2008JG000781CrossRefGoogle Scholar
  83. Sparks TH, Carey PD (1995) The responses of species to climate over two centuries: an analysis of the Marshman phenological record, 1736–1947. J Ecol 83:321–329CrossRefGoogle Scholar
  84. Sparks TH, Menzel A (2002) Observed changes in seasons: an overview. Int J Climate 22:1715–1725CrossRefGoogle Scholar
  85. Sparks TH, Tryjanowski P (2005) The detection of climate change impacts: some methodological considerations. Int J Climate 25:271–277CrossRefGoogle Scholar
  86. Studer S, Appenzeller C, Defila C (2005) Inter-annual variability and decadal trends in alpine spring phenology: a multivariate approach. Clim Change 73:395– 414CrossRefGoogle Scholar
  87. Studer S, Stöckli R, Appenzeller C et al. (2007) A comparative study of satellite and ground-based phenology. Int J Biometeorol 51:405–414CrossRefPubMedGoogle Scholar
  88. SSA-MTMGroup (2000) SSA-MTM toolkit 4.1 user’s guide. University of California, Los AngelesGoogle Scholar
  89. Tatli H, Dalfes HN, Mentes SS (2005) Surface air temperature variability over Turkey and its connection to large-scale upper air circulation via multivariate techniques. Int J Climatol 25:3 331–350CrossRefGoogle Scholar
  90. Thompson JD (1982) Spectrum estimation and harmonic analysis. Proc IEEE 70:1055–1096CrossRefGoogle Scholar
  91. Vautard R, Ghil M (1989) Singular spectrum analysis in nonlinear dynamics, with applications to paleoclimatic time series. Physica D 35:395–424CrossRefGoogle Scholar
  92. Vautard R, Yiou P, Ghil M (1992) Singular-spectrum analysis: a toolkit for short, noisy chaotic signals. Physica D 58:95–126CrossRefGoogle Scholar
  93. Weedon GP (2003) Time-series analysis and cyclostratigraphy: examining stratigraphic records of environmental cycles. Cambridge University Press, CambridgeCrossRefGoogle Scholar
  94. Wells K (2000) Long term cyclic and environmentally induced effects on flowering of four Box-ironbark eucalypts. Dissertation, University of MelbourneGoogle Scholar
  95. Wilson J (2002) Flowering ecology of a Box-ironbark Eucalyptus community. Dissertation, Deakin UniversityGoogle Scholar
  96. Wilson J, Bennett AF (1999) Patchiness of a floral resource: flowering of red ironbark Eucalyptus tricarpa in a box and ironbark forest. Victorian Nat 116:48–53Google Scholar
  97. White MA, Hoffman F, Hargrove WW (2005) A global framework for monitoring phenological responses to climate change. Geophys Res Lett 32(L04705)CrossRefGoogle Scholar
  98. Yiou P, Sornette D, Ghil M (2000) Data-adaptive wavelets and multi-scale singular-spectrum analysis. Physica D 142:254–290CrossRefGoogle Scholar
  99. Zhang X, Friedl MA, Schaaf CB et al. (2004) Climate controls on vegetation phonological patterns in northern mid- and high latitudes inferred from MODIS data. Glob Change Biol 10:1133–1145CrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media B.V. 2010

Authors and Affiliations

  1. 1.School of Mathematics and Statistics, University of South AustraliaAdelaideAustralia
  2. 2.Institute for Sustainable Systems and Technologies, University of South AustraliaMawson LakesAustralia
  3. 3.Department of Forest and Ecosystem ScienceUniversity of MelbourneVictoriaAustralia

Personalised recommendations