Download Microsoft PowerPoint - poster_discontinuit

PIECEWISE AND STRUCCHANGE: TEST OF TWO METHODS OF
CHANGE POINT ANALYSIS IN AGROCLIMATOLOGY
A.Chiaudani1, A. Berti2, M.Borin2, L.Mariani3
1. Veneto Region Environmental Protection Agency (ARPAV), Meteorological Centre of Teolo, [email protected]
2. Padua University, Department of Environmental Agronomy and Vegetable production, [email protected]; [email protected]
3. Milan University, Department of Vegetable production, [email protected]
Agency for Environmental Prevention
and Protection of Veneto Region
Introduction
In the last decades the perception to live a peculiar climatic phase has increased; this justify the importance of quantitative methods useful to describe the characters of this phase.
The temporal variability of historical datasets is frequently analyzed as a continuous process, described by linear interpolation methods. Nevertheless an important aspect of the evolution of
climate is represented by abrupt changes with different phases separated by breakpoints as shown for example by Sneyers, Palmieri and Siani (1993) and Lockwood (2001).
Each breakpoint splits the time-series in homogeneous sub-series whose trends must be analysed separately.
Some methods for discontinuity analysis are discussed in this work. A specific advantage of the discontinuity analyses is to focus on the climatic and non-climatic origin of the breakpoints
(e.g. changes in the circulation pattern behaviour at different scales, changes in land use, changes in the instruments and observation methods).
Methods and materials
An homogeneous archive for the 1956-2004 period was obtained from the analysis of daily data of precipitation and air temperature (source: National Hydro graphic Service, Meteorological
Centre of Teolo). These data were analysed by means of some different statistical methods as the simple linear regression, the mobile average, the piecewise linear (Tom and Miranda,
2004) and flat-steps (Bai and Perron, 2003) discontinuity methods.
Flat step e piecewise: conceptual definition
Piecewise and Flat Steps breakpoints in agriculture: maize, and wine grape proxy data
Yearly average max Summer Temperature in Veneto
Period 1956-2004. 9 stations
Yearly average max Summer Temperature
Yearly average max
Temperature in Veneto.
Period 1956-2004. 9 stations
1991
1980
1989
1978
The 9 mechanical
thermometric stations
Maize ripening date
Maize ripening date, class 600. Sowed the1st April
GDD: 1750 (BT= 8°C)
Flat-step: aims to identify the transition from a climatic homogeneous
phase to a new one (climatic normal), separating stationary sub-periods,
defined by their average values;
MAX T.. SUMMER
1991
Piecewise regression: aims to identify the initial changing moment of
the system, separating stationary sub-periods, defined by their trends;
TMAX PRIMAVERILI
The piecewise regression breakpoint, identifying the beginning of the
change, anticipates the flat-step breakpoint that identifies the moment
where the passage to a new stationary phase is demonstrable
Yearly average max Spring Temperature
MASSIMA
MEDIA
PRIMAVERILE in Veneto
YearlyTEMPERATURA
average max
Spring
Temperature
PERIODO 1956-2004
Period
1956-2004. 9 stations
20
19
18
The 49 mechanical raingauges
°C
17
16
15
14
1988
Winter Season and North Atlantic Oscillation (NAO)
Yearly average Winter Precipitation
2004
2002
2000
1998
1996
1994
1992
1990
1988
1986
1984
1982
1980
1978
1976
1974
1972
1970
1968
1966
1964
1962
1960
1958
1956
13
1981
Merlot Grape wine flowering
Merlot Data
Grape
wine flowering
in Conegliano
Fioritura
Merlot adate
Conegliano
Period 1964-2004
Yearly average Winter Precipitation in Veneto
Period 1956-2004. 9 stations
MAX TEMP. SPRING
200 mm
1991
Yearly average Precipitation
1956-1987
Yearly average Precipitation
1988-2004
Cortina (1180m). Winter days with soil snow
Other Proxy data
1981
1987
Annual anomaly of average Temp. in Italy (1961-2004)
and linear piecewise. Source: Toreti e Desiato (2007).
Arabba-ARPAV
Oak tree sprouting in Great Britain
Elaborations on data in
Maracchi e Baldi, 2006
1989
Yearly average max Winter Temperature
Yearly average max Winter Temperature in Veneto
Period 1956-2004. 9 stations
p<0.01
left: cloudiness trend at Atlantic mean latitudes (Huang et al. 2006). right: possible
breakpoint of cloudiness at Atlantic mean latitudes
1988
Starting of vegetative season in Germany
(fruit trees and field crops)
Elaborations on data in
Chmielewski F.M. et al., 2004
NAO index winter
Number of föhn days at Locarno-Monti (CH)
Period 1956-2003
Elaborations on data in Ambrosetti et al., 2005
1964
1986
Negative NAO index: climate more
cool and rainy
Positive NAO index: climate more
warm and dry
…. an outlier test ....identifies the decade
1981–1990 as the onset of climate change in
the North Atlantic:European sector with a trend
commencing in the beginning of the 1970s
Werner et al. 2000
Snow level in Friuli Venezia Giulia (cm)
Period : 1972-2006. December-April
elaborations on data belonging to Friuli Venezia Giulia Region
Stations:Forni di sopra (900m) Claut (620) Prescudin (Barcis)(540) Tarvisio (750)
Average temperature at Valley – (GB)
Period 1931-2000
Elaborations on data in
data set ECAD
Conclusions
The adoption of different methods for trend and discontinuity analysis can give complementary information useful in order to analyze the behavior of the time series which can be subdivided
in climatic and non climatic ones.
The flat-steps approach can be particularly interesting for agro-climatic purposes, because it permits to identify the reference period for climatic normal which are the basis for a lot of
evaluations about climatic change and adaptation of agricultural systems.
Bibliography
Bai J., Perron P. (2003), Computation and Analysis of Multiple Structural Change Models, Journal of Applied Econometrics, 18, 1-22.
Lockwood J.G., (2001), Abrupt and sudden climatic transitions and fluctuations; a review, Int. J. Climatol., 21,1153-1179.
Rial J.A., Pielke R., Beniston M., Claussen M., Canadell J., Cox P., Held H., Noblet-Ducoudré N., Prinn R., Reynolds F., Salas J., (2004). Non linearities, feedbacks and critical thresholds
within the earth’s climate system. Climatic Change, 65, 11–38.
Tom A.R., Miranda P.M.A. (2004), Piecewise linear fitting and trend changing points of climate parameters. Geophys Res Lett 31: L02207
Werner, P. C.,Gerstengarbe F.W.,Fraedrich K,Oesterle K.,2000.Recent climate change in the North Atlantic/European sector,International Journal of Climatology, Vol. 20, Issue 5, 2000:
463-471..