Advertisement

Modelling random uncertainty of eddy covariance flux measurements

  • Domenico VitaleEmail author
  • Massimo Bilancia
  • Dario Papale
Original Paper

Abstract

The eddy-covariance (EC) technique is considered the most direct and reliable method to calculate flux exchanges of the main greenhouse gases over natural ecosystems and agricultural fields. The resulting measurements are extremely important to characterize ecosystem exchanges of carbon, water, energy and other trace gases, and are widely used to validate or constrain parameter of land surface models via data assimilation techniques. For this purpose, the availability of both complete half-hourly flux time series and its associated uncertainty is mandatory. However, uncertainty estimation for EC data is challenging because the standard procedures based on repeated sampling are not suitable for this kind of measurements, and the presence of missing data makes it difficult to build any sensible time series model with time-varying second-order moments that can provide estimates of total random uncertainty. To overcome such limitations, this paper describes a new method in the context of the strategy based on the model residual approach proposed by Richardson et al. (Agric For Meteorol 148(1): 38–50, 2008). The proposed approach consists in (1) estimating the conditional mean process as representative of the true signal underlying observed data and (2) estimating the conditional variance process as representative of the total random uncertainty affecting EC data. The conditional mean process is estimated through the multiple imputation algorithm recently proposed by Vitale et al. (J Environ Inform  https://doi.org/10.3808/jei.201800391, 2018). The conditional variance process is estimated through the stochastic volatility model introduced by Beltratti and Morana (Econ Notes 30(2): 205–234, 2001). This strategy is applied to ten sites that are part of FLUXNET2015 dataset, selected in such a way to cover various ecosystem types under different climatic regimes around the world. The estimated uncertainty is compared with estimates by other well-established methods, and it is demonstrated that the scaling relationship between uncertainty and flux magnitude is preserved. Additionally, the proposed strategy allows obtaining a complete half-hourly time series of uncertainty estimates, which are expected to be useful for many users of EC flux data.

Keywords

Eddy covariance Net ecosystem exchange Global warming Uncertainty Conditional heteroskedasticity Time series Ecology 

Notes

Acknowledgements

Domenico Vitale (DV), Massimo Bilancia (MB), Dario Papale (DP). This paper has been started in the context of the ICOS-INWIRE Research Project funded by the European Community’s 7th Framework Program (FP7 / 2007-2013) under the Agreement \(\hbox {n}^{\circ }\) 313169 and continued and finalized under the ENVRIPLUS H2020 European Project (Grant Agreement 654182), that the authors thank for the support. DV conceived the study; DV and DP contributed to the study design; DV, MB and DP wrote the first draft of the manuscript. All authors equally contributed to the writing of Sect. 1. DV wrote Sects. 3.3456.3 and cared about the overall paper structure; MB wrote Sects. 3.13.26.16.2; Sects. 2 and 7 were written jointly by DV, MB and DP. All authors reviewed and revised the manuscript, approved the final version, and agreed to submit the manuscript for publication. This work used eddy covariance data acquired and shared by the FLUXNET community, including these networks: AmeriFlux, AfriFlux, AsiaFlux, CarboAfrica, CarboEuropeIP, CarboItaly, CarboMont, ChinaFlux, Fluxnet Canada, GreenGrass, ICOS, KoFlux, LBA, NECC, OzFlux TERN, TCOS Siberia, and USCCC. The ERA-Interim reanalysis data are provided by ECMWF and processed by LSCE. The FLUXNET eddy covariance data processing and harmonization was carried out by the European Fluxes Database Cluster, AmeriFlux Management Project, and Fluxdata project of FLUXNET, with the support of CDIAC and ICOS Ecosystem Thematic Center, and the OzFlux, ChinaFlux and AsiaFlux offices. The authors thank in particular the PIs that shared the data used in the study.

Supplementary material

477_2019_1664_MOESM1_ESM.pdf (23.3 mb)
Supplementary material 1 (pdf 23813 KB)

References

  1. Aubinet M, Vesala T, Papale D (eds) (2012) Eddy covariance. Springer, Dordrecht.  https://doi.org/10.1007/978-94-007-2351-1 Google Scholar
  2. Baillie RT, Bollerslev T (1992) Prediction in dynamic models with time-dependent conditional variances. J Econ 52(1–2):91–113.  https://doi.org/10.1016/0304-4076(92)90066-Z Google Scholar
  3. Baldocchi DD (2003) Assessing the eddy covariance technique for evaluating carbon dioxide exchange rates of ecosystems: past, present and future. Glob Change Biol 9(4):479–492.  https://doi.org/10.1046/j.1365-2486.2003.00629.x Google Scholar
  4. Baldocchi D, Falge E, Gu L, Olson R, Hollinger D, Running S, Anthoni P, Bernhofer C, Davis K, Evans R, Fuentes J, Goldstein A, Katul G, Law B, Lee X, Malhi Y, Meyers T, Munger W, Oechel W, Paw UKT, Pilegaard K, Schmid HP, Valentini R, Verma S, Vesala T, Wilson K, Wofsy S (2001) FLUXNET: a new tool to study the temporal and spatial variability of ecosystem-scale carbon dioxide, water vapor, and energy flux densities. Bull Am Meteorol Soc 82(11):2415–2434.  https://doi.org/10.1175/1520-0477(2001)082 Google Scholar
  5. Beltratti A, Morana C (2001) Deterministic and stochastic methods for estimation of intra-day seasonal components with high frequency data. Econ Notes 30(2):205–234.  https://doi.org/10.1111/j.0391-5026.2001.00054.x Google Scholar
  6. Beringer J, Hutley LB, Tapper NJ, Cernusak LA (2007) Savanna fires and their impact on net ecosystem productivity in North Australia. Glob Change Biol 13(5):990–1004.  https://doi.org/10.1111/j.1365-2486.2007.01334.x Google Scholar
  7. Billio M, Sartore D (2005) Stochastic volatility models: a survey with applications to option pricing and value at risk. In: Applied quantitative methods for trading and investment, John Wiley & Sons, Ltd, Chichester, UK, pp 239–291.  https://doi.org/10.1002/0470013265.ch8
  8. Blackwell M, Honaker J, King G (2015) A unified approach to measurement error and missing data: overview and applications. Soc Methods Res.  https://doi.org/10.1177/0049124115585360 Google Scholar
  9. Bollerslev T (1986) Generalized autoregressive conditional heteroskedasticity. J Econ 31(3):307–327.  https://doi.org/10.1016/0304-4076(86)90063-1 Google Scholar
  10. Bonal D, Bosc A, Ponton S, Goret JY, Burban B, Gross P, Bonnefond J, Elbers J, Longdoz B, Epron D et al (2008) Impact of severe dry season on net ecosystem exchange in the neotropical rainforest of French Guiana. Glob Change Biol 14(8):1917–1933.  https://doi.org/10.1111/j.1365-2486.2008.01610.x Google Scholar
  11. Bonan GB (2008) Forests and climate change: forcings, feedbacks, and the climate benefits of forests. Science 320(5882):1444–1449.  https://doi.org/10.1126/science.1155121 Google Scholar
  12. Boudt K, Cornelissen J, Payseur S, Nguyen G, Schermer M (2018) highfrequency: tools for highfrequency data analysis. https://cran.r-project.org/package=highfrequency. Accessed 24 Jan 2019
  13. Dempster AP, Laird NM, Rubin DB (1977) Maximum likelihood from incomplete data via the EM algorithm. J R Stat Soc Ser B 39(1):1–38Google Scholar
  14. Efron B (1994) Missing data, imputation, and the bootstrap. J Am Stat Assoc 89(426):463–475.  https://doi.org/10.1080/01621459.1994.10476768 Google Scholar
  15. Efron B (2012) Bayesian inference and the parametric bootstrap. Ann Appl Stat 6(4):1971–1997.  https://doi.org/10.1214/12-AOAS571 Google Scholar
  16. Falge E, Baldocchi D, Olson R, Anthoni P, Aubinet M, Bernhofer C et al (2001) Gap filling strategies for defensible annual sums of net ecosystem exchange. Agric For Meteorol 107(1):43–69.  https://doi.org/10.1016/S0168-1923(00)00225-2 Google Scholar
  17. Fathian F, Fakheri-Fard A, Modarres R, van Gelder PHAJM (2018) Regional scale rainfall–runoff modeling using VARX–MGARCH approach. Stoch Environ Res Risk Assess 32(4):999–1016.  https://doi.org/10.1007/s00477-017-1428-6 Google Scholar
  18. Finkelstein PL, Sims PF (2001) Sampling error in eddy correlation flux measurements. J Geophys Res Atmos 106(D4):3503–3509.  https://doi.org/10.1029/2000JD900731 Google Scholar
  19. Foken T, Leuning R, Oncley SR, Mauder M, Aubinet M (2012) Corrections and data quality control. In: Aubinet M, Vesala T, Papale D (eds) Eddy covariance. Springer, Dordrecht, pp 85–131.  https://doi.org/10.1007/978-94-007-2351-1_4 Google Scholar
  20. Gelman A, Carlin J, Stern H, Dunson D, Vehtari A, Rubin D (2013) Bayesian data analysis, 3rd edn. Texts in statistical science. Chapman and Hall/CRC, LondonGoogle Scholar
  21. Giller GL (2005) A generalized error distribution. SSRN Electron J.  https://doi.org/10.2139/ssrn.2265027 Google Scholar
  22. Goodnight JH (1979) A tutorial on the SWEEP operator. Am Stat 33(3):149.  https://doi.org/10.2307/2683825 Google Scholar
  23. Hamilton JD (1994) Time series analysis. Princeton University Press, PrincetonGoogle Scholar
  24. Harvey AC (1989) Forecasting, structural time series models and the Kalman filter. Cambridge University Press, CambridgeGoogle Scholar
  25. Hollinger DY, Richardson AD (2005) Uncertainty in eddy covariance measurements and its application to physiological models. Tree Physiol 25(7):873–885.  https://doi.org/10.1093/treephys/25.7.873 Google Scholar
  26. Hollinger DY, Aber J, Dail B, Davidson EA, Goltz SM, Hughes H, Leclerc MY, Lee JT, Richardson AD, Rodrigues C, Scott N, Achuatavarier D, Walsh J (2004) Spatial and temporal variability in forest-atmosphere CO\(_2\) exchange. Glob Change Biol 10(10):1689–1706.  https://doi.org/10.1111/j.1365-2486.2004.00847.x Google Scholar
  27. Honaker J, King G (2010) What to do about missing values in time-series cross-section data. Am J Polit Sci 54(2):561–581.  https://doi.org/10.1111/j.1540-5907.2010.00447.x Google Scholar
  28. Honaker J, King G, Blackwell M (2011) Amelia II: a program for missing data. J Stat Softw 45(7):1–47.  https://doi.org/10.18637/jss.v045.i07 Google Scholar
  29. Houghton R (2005) Aboveground forest biomass and the global carbon balance. Glob Change Biol 11(6):945–958Google Scholar
  30. Hui D, Wan S, Su B, Katul G, Monson R, Luo Y (2004) Gap-filling missing data in eddy covariance measurements using multiple imputation (MI) for annual estimations. Agric For Meteorol 121(1–2):93–111.  https://doi.org/10.1016/S0168-1923(03)00158-8 Google Scholar
  31. Huisman R, Huurman C, Mahieu R (2007) Hourly electricity prices in day-ahead markets. Energy Econ 29(2):240–248.  https://doi.org/10.1016/j.eneco.2006.08.005 Google Scholar
  32. Imai K, King G, Lau O (2008) Toward a common framework for statistical analysis and development. J Comput Graph Stat 17(4):892–913.  https://doi.org/10.1198/106186008X384898 Google Scholar
  33. Kastner G (2016) Dealing with stochastic volatility in time series using the R package stochvol. J Stat Softw 69(5):1–30.  https://doi.org/10.18637/jss.v069.i05 Google Scholar
  34. Kato T, Knorr W, Scholze M, Veenendaal E, Kaminski T, Kattge J, Gobron N (2013) Simultaneous assimilation of satellite and eddy covariance data for improving terrestrial water and carbon simulations at a semi-arid woodland site in Botswana. Biogeosciences 10(2):789–802.  https://doi.org/10.5194/bg-10-789-2013 Google Scholar
  35. Kotz S, Kozubowski TJ, Podgórski K (2001) The Laplace distribution and generalizations. Birkhäuser Boston, Boston.  https://doi.org/10.1007/978-1-4612-0173-1 Google Scholar
  36. Kristensen D (2010) Nonparametric filtering of the realized spot volaitlity: a kernel-based approach. Econ Theory 26(1):60–93.  https://doi.org/10.1017/S0266466609090616 Google Scholar
  37. Lasslop G, Reichstein M, Kattge J, Papale D (2008) Influences of observation errors in eddy flux data on inverse model parameter estimation. Biogeosciences 5(5):1311–1324.  https://doi.org/10.5194/bg-5-1311-2008 Google Scholar
  38. Lee KJ, Carlin JB (2010) Multiple imputation for missing data: fully conditional specification versus multivariate normal imputation. Am J Epidemiol 171(5):624–632.  https://doi.org/10.1093/aje/kwp425 Google Scholar
  39. Lenschow DH, Kristensen L (1985) Uncorrelated noise in turbulence measurements. J Atmos Ocean Technol 2(1):68–81.  https://doi.org/10.1175/1520-0426(1985) Google Scholar
  40. Little RJA, Rubin DB (2002) Statistical analysis with missing data, 2nd edn. Wiley, Hoboken.  https://doi.org/10.1002/9781119013563 Google Scholar
  41. Lomb NR (1976) Least-squares frequency analysis of unequally spaced data. Astrophys Space Sci 39(2):447–462Google Scholar
  42. Mandelbrot B (1963) The variation of certain speculative prices. J Bus 36(4):394.  https://doi.org/10.1086/294632 Google Scholar
  43. Mauder M, Foken T (2004) Documentation and instruction manual of the eddy covariance software package TK2. University of Bayreuth, Abt Mikrometeorol, Bayreuth, pp 26–42Google Scholar
  44. Mauder M, Cuntz M, Drüe C, Graf A, Rebmann C, Schmid HP, Schmidt M, Steinbrecher R (2013) A strategy for quality and uncertainty assessment of long-term eddy-covariance measurements. Agric For Meteorol 169:122–135.  https://doi.org/10.1016/j.agrformet.2012.09.006 Google Scholar
  45. Meyer W, Kondrlovà E, Koerber G (2015) Evaporation of perennial semi-arid woodland in southeastern Australia is adapted for irregular but common dry periods. Hydrol Process 29(17):3714–3726.  https://doi.org/10.1002/hyp.10467 Google Scholar
  46. Moffat AM, Papale D, Reichstein M, Hollinger DY, Richardson AD, Barr AG, Beckstein C, Braswell BH, Churkina G, Desai AR, Falge E, Gove JH, Heimann M, Hui D, Jarvis AJ, Kattge J, Noormets A, Stauch VJ (2007) Comprehensive comparison of gap-filling techniques for eddy covariance net carbon fluxes. Agric For Meteorol 147(3–4):209–232.  https://doi.org/10.1016/j.agrformet.2007.08.011 Google Scholar
  47. Moncrieff J, Malhi Y, Leuning R (1996) The propagation of errors in long-term measurements of land-atmosphere fluxes of carbon and water. Glob Change Biol 2(3):231–240.  https://doi.org/10.1111/j.1365-2486.1996.tb00075.x Google Scholar
  48. Pan Y, Birdsey RA, Fang J, Houghton R, Kauppi PE, Kurz WA, Phillips OL, Shvidenko A, Lewis SL, Canadell JG, Ciais P, Jackson RB, Pacala SW, McGuire AD, Piao S, Rautiainen A, Sitch S, Hayes D (2011) A large and persistent carbon sink in the world’s forests. Science 333(6045):988–993.  https://doi.org/10.1126/science.1201609 Google Scholar
  49. Papale D, Valentini R (2003) A new assessment of European forests carbon exchanges by eddy fluxes and artificial neural network spatialization. Glob Change Biol 9(4):525–535.  https://doi.org/10.1046/j.1365-2486.2003.00609.x Google Scholar
  50. Papale D, Reichstein M, Aubinet M, Canfora E, Bernhofer C, Kutsch W, Longdoz B, Rambal S, Valentini R, Vesala T, Yakir D (2006) Towards a standardized processing of net ecosystem exchange measured with eddy covariance technique: algorithms and uncertainty estimation. Biogeosciences 3(4):571–583.  https://doi.org/10.5194/bg-3-571-2006 Google Scholar
  51. Park J, Byun K, Choi M, Jang E, Lee J, Lee Y, Jung S (2015) Evaluation of statistical gap fillings for continuous energy flux (evapotranspiration) measurements for two different land cover types. Stoch Environ Res Risk Assess 29(8):2021–2035.  https://doi.org/10.1007/s00477-015-1101-x Google Scholar
  52. Pilegaard K, Ibrom A, Courtney MS, Hummelshøj P, Jensen NO (2011) Increasing net CO\(_2\) uptake by a danish beech forest during the period from 1996 to 2009. Agric For Meteorol 151(7):934–946.  https://doi.org/10.1016/j.agrformet.2011.02.013 Google Scholar
  53. Rambal S, Joffre R, Ourcival J, Cavender-Bares J, Rocheteau A (2004) The growth respiration component in eddy CO\(_2\) flux from a Quercus ilex mediterranean forest. Glob Change Biol 10(9):1460–1469.  https://doi.org/10.1111/j.1365-2486.2004.00819.x Google Scholar
  54. Rannik Ü, Peltola O, Mammarella I (2016) Random uncertainties of flux measurements by the eddy covariance technique. Atmos Meas Tech 9(10):5163–5181.  https://doi.org/10.5194/amt-9-5163-2016 Google Scholar
  55. Reichstein M, Falge E, Baldocchi D, Papale D (2005) On the separation of net ecosystem exchange into assimilation and ecosystem respiration: review and improved algorithm. Glob Change Biol 11:1–16.  https://doi.org/10.1111/j.1365-2486.2005.001002.x Google Scholar
  56. Richardson AD, Hollinger DY (2005) Statistical modeling of ecosystem respiration using eddy covariance data: maximum likelihood parameter estimation, and Monte Carlo simulation of model and parameter uncertainty, applied to three simple models. Agric For Meteorol 131(3–4):191–208.  https://doi.org/10.1016/j.agrformet.2005.05.008 Google Scholar
  57. Richardson AD, Mahecha MD, Falge E, Kattge J, Moffat AM, Papale D, Reichstein M, Stauch VJ, Braswell BH, Churkina G, Kruijt B, Hollinger DY (2008) Statistical properties of random CO\(_2\) flux measurement uncertainty inferred from model residuals. Agric For Meteorol 148(1):38–50.  https://doi.org/10.1016/j.agrformet.2007.09.001 Google Scholar
  58. Richardson AD, Aubinet M, Barr AG, Hollinger DY, Ibrom A, Lasslop G, Reichstein M (2012) Uncertainty quantification. In: Aubinet M, Vesala T, Papale D (eds) Eddy covariance. Springer, Dordrecht, pp 173–209.  https://doi.org/10.1007/978-94-007-2351-1_7 Google Scholar
  59. Rubin DB (1987) Multiple imputation for nonresponse in surveys. Wiley series in probability and statistics. Wiley, Hoboken.  https://doi.org/10.1002/9780470316696 Google Scholar
  60. Rubin DB (1996) Multiple imputation after 18+ years. J Amn Stat Assoc 91(434):473–489.  https://doi.org/10.2307/2291635 Google Scholar
  61. Sabbatini S, Arriga N, Bertolini T, Castaldi S, Chiti T, Consalvo C, Njakou Djomo S, Gioli B, Matteucci G, Papale D (2016) Greenhouse gas balance of cropland conversion to bioenergy poplar short-rotation coppice. Biogeosciences 13(1):95–113.  https://doi.org/10.5194/bg-13-95-2016 Google Scholar
  62. Scargle JD (1982) Studies in astronomical time series analysis. II-statistical aspects of spectral analysis of unevenly spaced data. Astrophys J 263:835–853Google Scholar
  63. Schafer J (1997) Analysis of incomplete multivariate data, vol 72. Monographs on statistics & applied probability. Chapman & Hall, Boca Raton.  https://doi.org/10.1201/9781439821862 Google Scholar
  64. Schafer JL, Graham JW (2002) Missing data: our view of the state of the art. Psychol Methods 7(2):147–177.  https://doi.org/10.1037/1082-989X.7.2.147 Google Scholar
  65. Sénégas J, Wackernagel H, Rosenthal W, Wolf T (2001) Error covariance modeling in sequential data assimilation. Stoch Environ Res Risk Assess 15(1):65–86.  https://doi.org/10.1007/PL00009788 Google Scholar
  66. Shephard N (1996) Statistical aspects of ARCH and stochastic volatility. In: Cox DR, Hinkley DV, Barndorff-Nielsen OE (eds) Time series models in econometrics, finance and other fields. Chapman & Hall, London, pp 1–67Google Scholar
  67. Stauch VJ, Jarvis AJ (2006) A semi-parametric gap-filling model for eddy covariance CO\(_2\) flux time series data. Glob Change Biol 12(9):1707–1716.  https://doi.org/10.1111/j.1365-2486.2006.01227.x Google Scholar
  68. Stauch VJ, Jarvis AJ, Schulz K (2008) Estimation of net carbon exchange using eddy covariance CO\(_2\) flux observations and a stochastic model. J Geophys Res 113(D3):D03101.  https://doi.org/10.1029/2007JD008603 Google Scholar
  69. Sulman B, Desai A, Cook B, Saliendra N, Mackay D (2009) Contrasting carbon dioxide fluxes between a drying shrub wetland in northern Wisconsin, USA, and nearby forests. Biogeosciences 6(6):1115–1126.  https://doi.org/10.5194/bg-6-1115-2009 Google Scholar
  70. Suni T, Rinne J, Reissell A, Altimir N, Keronen P, Rannik U, Maso M, Kulmala M, Vesala T (2003) Long-term measurements of surface fluxes above a scots pine forest in Hyytiala, southern Finland, 1996–2001. Boreal Environ Res 8(4):287–302Google Scholar
  71. Taylor SJ (1994) Modeling stochastic volatility: a review and comparative study. Math Finance 4(2):183–204.  https://doi.org/10.1111/j.1467-9965.1994.tb00057.x Google Scholar
  72. Tzikas D, Likas A, Galatsanos N (2008) The variational approximation for Bayesian inference. IEEE Signal Process Mag 25(6):131–146.  https://doi.org/10.1109/MSP.2008.929620 Google Scholar
  73. van Buuren S (2012) Flexible imputation of missing data, vol 20125245. Interdisciplinary statistics series. Chapman and Hall/CRC, Boca Raton.  https://doi.org/10.1201/b11826 Google Scholar
  74. Verma SB, Dobermann A, Cassman KG, Walters DT, Knops JM, Arkebauer TJ, Suyker AE, Burba GG, Amos B, Yang H et al (2005) Annual carbon dioxide exchange in irrigated and rainfed maize-based agroecosystems. Agric Fort Meteorol 131(1):77–96.  https://doi.org/10.1016/j.agrformet.2005.05.003 Google Scholar
  75. Vitale D, Bilancia M, Papale D (2018) A multiple imputation strategy for eddy covariance data. J Environ Inform.  https://doi.org/10.3808/jei.201800391 Google Scholar
  76. Wohlfahrt G, Hammerle A, Haslwanter A, Bahn M, Tappeiner U, Cernusca A (2008) Seasonal and inter-annual variability of the net ecosystem CO\(_2\) exchange of a temperate mountain grassland: effects of weather and management. J Geophys Res Atmos.  https://doi.org/10.1029/2007jd009286 Google Scholar
  77. Wood AM, White IR, Royston P (2008) How should variable selection be performed with multiply imputed data? Stat Med 27(17):3227–3246.  https://doi.org/10.1002/sim.3177 Google Scholar
  78. Wu CFJ (1983) On the convergence properties of the EM algorithm. Ann Stat 11(1):95–103.  https://doi.org/10.1214/aos/1176346060 Google Scholar
  79. Ye M, Meyer PD, Lin YF, Neuman SP (2010) Quantification of model uncertainty in environmental modeling. Stoch Environ Res Risk Assess 24(6):807–808.  https://doi.org/10.1007/s00477-010-0377-0 Google Scholar
  80. Zheng Y, Han F (2016) Markov Chain Monte Carlo (MCMC) uncertainty analysis for watershed water quality modeling and management. Stoch Environ Res Risk Assess 30(1):293–308.  https://doi.org/10.1007/s00477-015-1091-8 Google Scholar

Copyright information

© Springer-Verlag GmbH Germany, part of Springer Nature 2019

Authors and Affiliations

  1. 1.Department for Innovation in Biological, Agro-Food and Forest Systems (DIBAF)University of TusciaViterboItaly
  2. 2.Ionian Department of Law, Economics and EnvironmentUniversity of Bari Aldo MoroTarantoItaly

Personalised recommendations