Note sul progetto IPHEN

Luigi Mariani Università degli Studi di Milano - Di.Pro.Ve.

luigi.mariani@unimi.it

IPHEN project

a co-operative project finalised to phenological a co-operative project finalised to phenological monitoring and modelling with regular production and monitoring and modelling with regular production and

broadcast of weekly phenological maps broadcast of weekly phenological maps

IPHEN was promoted by a group of Italian phenologists during a workshop held in Rome (14-15 dec 2005) and the first products

were issued in March 2006

Important tools to evaluate the spatial variability of phenomena, giving a territorial view

Long tradition in phenological carthography based on manual methods (subjective criteria)

More recently: automatic methods (geostatistics)

Both manual and automatic methods should take into account the causal link between phenology and ancillary variables (height, aspect, etc.)

Phenological maps

The 1882 map of Lilac flowering produced by Hoffmann and Ihne

(Larcher, 1983)

List of participants for 2010

Co-ordinatorCo-ordinator – CRA-Cma (Research unit for meteo-climatology) – CRA-Cma (Research unit for meteo-climatology)

PartipantsPartipantsCra (Council for agricultural research)Cra (Council for agricultural research)- (CRA-CMA)- (CRA-CMA)- Centro di ricerca per la viticoltura (Conegliano TV) (CRA-VIT)- Centro di ricerca per la viticoltura (Conegliano TV) (CRA-VIT)- Centro di ricerca per l’olivicoltura e l’industria olearia (Rende) (CRA-OLI)- Centro di ricerca per l’olivicoltura e l’industria olearia (Rende) (CRA-OLI)- Unità di ricerca per l’uva da tavola e la vitivinicoltura in ambiente mediterraneo (Turi BA) (CRA-UTV)- Unità di ricerca per l’uva da tavola e la vitivinicoltura in ambiente mediterraneo (Turi BA) (CRA-UTV)

UniversitiesUniversities- Università degli Studi di Bologna - Facoltà di Agraria, Dip. di Biologia Evoluzionistica; Dip. di Produzione Vegetale, Dip. di Scienze - Università degli Studi di Bologna - Facoltà di Agraria, Dip. di Biologia Evoluzionistica; Dip. di Produzione Vegetale, Dip. di Scienze e Tecnologie Agroambientalie Tecnologie Agroambientali- Università Cattolica del Sacro Cuore - Università Cattolica del Sacro Cuore - Università degli Studi di Firenze Centro Interdipartimentale di Bioclimatologia CIBIC-UNIFI- Università degli Studi di Firenze Centro Interdipartimentale di Bioclimatologia CIBIC-UNIFI- Università degli Studi di Milano – Facoltà di Agraria Dipartimento di Produzione Vegetale- Università degli Studi di Milano – Facoltà di Agraria Dipartimento di Produzione Vegetale- Università degli Studi di Palermo (Motisi)- Università degli Studi di Palermo (Motisi)- Università degli Studi di Pisa - Centro Interdipartimentale di Ricerche Agro-Ambientali "Enrico Avanzi"- Università degli Studi di Pisa - Centro Interdipartimentale di Ricerche Agro-Ambientali "Enrico Avanzi"- Università degli Studi di Roma Tor Vergata – Dip. di Biologia- Università degli Studi di Roma Tor Vergata – Dip. di Biologia- Università degli Studi di Udine- Università degli Studi di Udine- Università degli Studi di Sassari- Università degli Studi di Sassari- Università degli Studi dell’Aquila – Dip. di Scienze Ambientali- Università degli Studi dell’Aquila – Dip. di Scienze Ambientali- Università degli Studi di Torino- Università degli Studi di Torino

List of 2010 participants

PartipantsPartipantsCNR – National Research Council)CNR – National Research Council)- Istituto di Biometeorologia Ibimet- Istituto di Biometeorologia Ibimet

AssociationsAssociations AIA - Associazione Italiana di Aerobiologia www.ilpolline.itAIA - Associazione Italiana di Aerobiologia www.ilpolline.itAIAM - Associazione Italiana di Agrometeorologia. www.agrometeorologia.itAIAM - Associazione Italiana di Agrometeorologia. www.agrometeorologia.itUnaapi, Unione Nazionale Associazioni Apicoltori Italiani www.mieliditalia.itUnaapi, Unione Nazionale Associazioni Apicoltori Italiani www.mieliditalia.it

AAgrometeorological and extension servicesgrometeorological and extension services- Regione Sicilia – Servizio Informativo Agrometeorologico Siciliano- Regione Sicilia – Servizio Informativo Agrometeorologico Siciliano- Regione Calabria - Arssa Calabria – Servizio Agrometeorologico- Regione Calabria - Arssa Calabria – Servizio Agrometeorologico- Regione Emilia Romagna - Arpa Emilia Romagna - Servizio Idrometeorologico- Regione Emilia Romagna - Arpa Emilia Romagna - Servizio Idrometeorologico- Regione Lazio – ARSIAL- Regione Lazio – ARSIAL- Fondazione Edmund Mach - S. Michele, Trento- Fondazione Edmund Mach - S. Michele, Trento- Consorzio di Difesa di Brindisi CODIVABRI- Consorzio di Difesa di Brindisi CODIVABRI- Regione Sardegna - Arpa Sardegna- Regione Sardegna - Arpa Sardegna- Regione Liguria – Ass.to Agricoltura, Settore Servizi alle Imprese Agricole C.A.A.R. (Centro di Agrometeorologia Applicata Reg.ale)- Regione Liguria – Ass.to Agricoltura, Settore Servizi alle Imprese Agricole C.A.A.R. (Centro di Agrometeorologia Applicata Reg.ale)- Regione Piemonte - Ass.to Agricoltura- Regione Piemonte - Ass.to Agricoltura- Regione Val d’Aosta - Institut Agricole Régional- Regione Val d’Aosta - Institut Agricole Régional- Amministrazione Provinciale di Pesaro ed Urbino (Ufficio Ambiente - Serv.4.3- Amministrazione Provinciale di Pesaro ed Urbino (Ufficio Ambiente - Serv.4.3

Iphen – SELECTED SPECIES

Reasons of this choice

Grapevine: economic relevance. Activities are referred to Chardonnay Cabernet s., international varieties present in the whole Italian area and respectively early and late varieties, “enveloping” the main part of Italian varieties.

Black elder: spontaneous, present in whole Italian area.

Locust tree: spontaneous, important for apiculture and production of honey

Olive tree: economic relevance. allergenic species.

Cypress (common and arizonica): ornamental species and allergenic species too.

Manuals with observational standard

Standard sheets for observational site characterization

Weekly observation sheets

DOCUMENTATION FOR OBSERVERS

PARSIMONY in input variables (meteo & pheno observations)

MINIMAL PARAMETERIZATION - the reduced availability of phenological series is a limit for calibration/validation of models

ROBUSTNESS – maps must be produced, also if observational datasets (meteo & pheno) are incomplete

FLEXIBILITY – adaptability to a wide set of species/varieties.

RESOLUTION – until meteorological mesoscale

STANDARDIZATION – BBCH scale

FORECAST POWER– not only analysis but also forecasts

The Iphen approach to phenological cartographyPrerequisites considered

IPHEN appoach main features derived from these prerequisites

Simple empirical models (based on NHH) with daily time step

A scheme based on observational data for model output correction

Weekly issues of analysis and forecast products.

Iphen – Timetable of map emissions -last year

→ Phenological rhythms of plants are ruled by a biological clock that is generally difficult to read (it is strictly related to hormones and plant physiology)

→ For this reason a thermal clock is often considered instead of the biological clock

Limitations: linearity of response is a pre-requsite for our thermometric sensors. Plants are not true thermometers (their biological objectives justify non linear responses to temperatures or the same answer to different thermal forcing).

The NHH approach

How to read the thermal clock of plants

Thermal Units: standard approach adopted from the epoch of Réaumur (XVIIIth century).

Normal Heat Hours (NHH): Hourly temperatures normalized by means of a response curve representing the effectiveness of temperature for plant phenology

Advantage of NHH: they overcome the main limitation of thermal units (overestimate of super-optimal temperatures).

How NHH work

Beta function (Wang & Engel,

1998)

An hour spent at 17 °C corresponds to 0.53 Normal Heat Hours (NHH)

- effetto latitudinale- effetto dell'orografia- effetto “Tirreno” (area tirrenica più calda di quella adriatica

alla stessa latitudine)

NHH–parameterization of the beta function for Iphen project

  CARDINAL TEMPERATURES

  min opt max

Vitis vinifera 7 26 35

Olea europaea 7 26 35

Robinia pseduoacacia 10 22 38

Sambucus nigra 1 20 32

Cupressus sempervirens 0 8 17

Phenological phaseCultivar Location Latitude

(°)Sprouting(BBCH 8)

Flowering(BBCH 65)

Veraison(BBCH 81)

Ripening(BBCH 89)

Cagliari 39 336 894 2053 2899Sigonella 37 335 885 1868 2349Firenze 44 258 890 1906 2549Ronchi 45 175 810 1906 2506

Chardonnay

Bergamo 45 121 764 1829 2521Cagliari 39 401 1005 2068 2966Sigonella 37 400 1004 2067 2965Firenze 44 323 927 1990 2888Ronchi 45 240 844 1907 2805

CabernetSauvignon

Bergamo 45 186 790 1853 2751

The change of perspective imposed by NHH approach

Chardonnay & Cabernet Sauvignon - NHH from 1 January (reference values for some Italian stations – 1961-1990 period)

Simple empirical equations are adopted

Examples (equations obtained working on Iphen data 2006-2007):

For Cabernet s.SCALE

comodo=(-0.978*Ygeo+55.319)*Ln(NHH)+7.360*Ygeo-374.450

For ChardonnaySCALE

comodo=(-0.916*Ygeo+52.092)*Ln(NHH)+7.069*Ygeo-355.869

Ygeo=latitude in geographical co-ordinates

Why SCALEcomodo

? It is an intermediate scale useful to correct

the non-linear aspects of the BBCH.

From NHH to BBCH phases

Transcoding table BBCH – Scalecomodo

Pre-flowering phases

Flowering and fruit development phases

IPHEN MODEL

Developed in Standard Pascal in Delphi environment

The whole process is automatized (meteorological data management, production of different information layers, match of observations vs. simulations, etc.)

The main products are:

ANALISYS Italian maps of phenological phases reached at the broadcast day

FORECASTS Italian maps of date (doy) when a given phenological stage will be reached.

ANALYSIS MODEL SCHEME – data sources

Pixel of about 2 km in geographical coord.

92 stations from Cra and National weather service

Phenogical observers

CARTA DI ANALISIANALYSIS MODEL SCHEME

Daily thermal fields (Tx, Tn)NHH cumulated from 1 Jan First guess map

Error mapFinal map

CARTE DI PREVISIONESCHEME OF THE FORECAST MODEL

Final analysis map (initial conditions) NHH cumulated from 1th Jan

Forecast maps of 61 and 65 BBCH NHH forecast from GFS and climatology

Some examples

CARTA DI ANALISIANALYSIS – Cupressus sempervirens L. – 5 March 2010

CARTE DI PREVISIONEFORECAST – Cupressus sempervirens L. – 5 March 2010

CARTE DI PREVISIONEFORECAST – Cupressus sempervirens L. – 5 March 2010

ANALYSIS – Robinia pseudoacacia L. – 13 May 2010

FORECAST – Robinia pseudoacacia L. – 13 May 2010

ANALYSIS - Olea europaea L. – 4 June 2010

ANALYSIS – Sambucus nigra L. – 6 May 2010

ANALYSIS – Grapevine cv Chardonnay – 20 May 2010

FORECAST – Grapevine cv Chardonnay – 20 May 2010

FORECAST – Grapevine cv Chardonnay – 20 May 2010

Phenology as proxy of meteo-climatic variability

Italy has a North-South extension of about 1200 km, ranging from Koeppen macroclimate Cfb (oceanic climate) to Csa (Mediterranean climate), with a complex relief and with a great variety of mesoclimates.In this context, phenological maps can be useful to appreciate a lot of different effects like:LATITUDINAL EFFECTSTOPOCLIMATIC EFFETTSSEA EFFECTS (e.g.: Thyrrenian areas often warmer of the Adriatic ones at the same latitude due to the outbreaks of cold continental air masses).EXCEPTIONS DUE TO METEOROLOGICAL ANOMALIES

OVERVIEW

Sambucus nigra - reproductive phase – 20 april 2006

Sambucus nigra – 11 july 2008

Overall scheme of general circulation

Annular circulation (30-60°)

Polar Vortex (60-90°)

Hadley cell (0-30°)

General circulation using clouds as tracers

Tropical area: a very active convective zone near the equatorMid latitudes: frontal systems with a West-East field of motion (annular circulation alias westerlies)

Annular circulation and energy exchange between low and high latitudes

Meteosat visibile 25 dicembre 2004 h 12 UTC

ARIA ARTICA

ARIA SUBTROPICALE

Mean state of annular circulation and blocking

The mean state of mid latitudes annular circulation is zonal(from west to east - slowly ondulated)

Blocking = deviations from this dominant regime due to cyclonic or anticyclonic areas that block the westerly flow,

Blokings can persist for days / weeks /months.

Blocking weather patterns

Omega block Inverse S shaped(Rex block)

Source: Jeff Haby - http://www.theweatherprediction.com/blocking/Cutoff High Cutoff low Split flow

Zonal (mean) regime

Summer floods in UK (2008)Dust bowl in USA (1935)

blockings climatology and extreme eventsBlockings climatology (frequency and persistence) is strictly related to severe weather anomalies like:- hot outbreaks (e.g.: European hot wave of 2003)- cold outbreaks (e.g.: European cold winters 1929,1956,1985)- persistent rainfalls (e.g.: Florence flloding of 1966)- droughts (eg: dust bowl of '30 years in Usa)

Boreal hemisphere: frequency of blockings

Using the blocking index of Tibaldi and Molteni (1990) we can consider the frequency of "blocked days" for each of the four traditional seasons (DJF, MAM, JJA, SON) for the northern hemisphere for the period 1950 to 2000.

Analysis carried out on NCEP Ncar reanalysissource: http://www.cpc.ncep.noaa.gov/products/precip/CWlink/blocking/seasonal_nh/seasonal_nh.shtml

Cabernet S. Chardonnay

31 july 2006

July 2006 – mean abs. Topography - 700 hPa(http://www.esrl.noaa.gov/psd/data/gridded/data.ncep.reanalysis.pressure.html)

Long time evolution of zonal regime the abrupt change of '80

Long time behaviour of annular circulation can be described by indexes bases on atmospheric pressure distributions.

AO and Nao: circulation indexes adopted to describe behaviour of zonal circulation in the North hemisphere.An abrupt AO (and Nao) change of phase was observed at the end of the '80 with strengthening of zonal regime.

http://www.jisao.washington.edu/ao/

NAO positive phases Effects on the Euro-Mediterranean area

Source - Trouet et al., 2009. Persistent Positive North Atlantic Oscillation Mode Dominated the Medieval Climate Anomaly, Science, 3 april 2009, Vol 324

Anomalies in SLP, temperarure and precipitation obtained from proxies and instrumental data referred to 1659-1995 period

Historical precedents

NAO recostruction from proxy data from Marocco, Alps and Scotland).

Trouet et al., 2009. Persistent Positive North Atlantic Oscillation Mode Dominated the Medieval Climate Anomaly, Science, 3 april 2009, Vol 324

Mean yearly temperature of 18 European stations (1951-2009)(available data: 93%)

With the 99% of confidence the breakpoint drops between 1983 and 1991. Most probable year=1987 (mean 1951-1987=6.9; mean 1988-2009=7.9)

[breakpoint analysis carried out with Struchange library – R cran]

1987

Oak budding in UK (1950-2005)(source: Maracchi, Baldi, 2006)

Con il 90% di probabilità la discontinuità ricade fra 1985 e 1989. Anno più probabile di discontinuità= 1987 (media 1950 -1986=117; media 1989-1997=99)

[analisi di discontinuità eseguita con la libreria statistica Struchange]

27 aprile

9 aprile

1987

Milano Linate

Precipitazioni 1921-2006 – Bordeaux e Francoforte (fonte: Ecad)

Precipitazione media annua 1950-2006 - 18 stazioni della provincia di Sondrio (Dati progetto RICLIC WARM)

Media 1950-1989 1238 mm/anno

media 1990-2006 1157 mm/anno

Milano Brera - Precipitazione annua 1764-2010 (in evidenza i 5 massimi assoluti)

Milano Brera - Precipitazione annua 1764-2010

Valori medi (mm/anno)

- Media 1764-1800=928 - Media 1801-1850=1053- Media 1851-1900= 1023- Media 1901-1950= 963- Media 1951-2000= 1016- Media 2001-2010= 938

→ Media 1764-2010=998

Temperatures (1979-2011)

Source: University of 'Alabama – Huntsville (satellite data MSU-UAH) numerical data available at: http://vortex.nsstc.uah.edu/public/msu/t2lt/uahncdc.lt

Perspectives

- effetto latitudinale- effetto dell'orografia- effetto “Tirreno” (area tirrenica più calda di quella adriatica

alla stessa latitudine)

- Variability of the dataset with relevant disequilibrium among different areas of the country

- Representativity of spatializations (problem of microscale effects like urban effects, topoclimatic effects and so on)

- extension to other relevant species

PROBLEMS

- effetto latitudinale- effetto dell'orografia- effetto “Tirreno” (area tirrenica più calda di quella adriatica

alla stessa latitudine)

IMPROVEMENT of phenological models by means of specific calibration and validation activities carried out on observational datasets gathered by the IPHEN network

REFINEMENT of the forecast runs

PORTING ATCIVITY (from the present prototypal software to an engineered product working into the national data processing system of CRA-Cma)

DOWNSCALINGmesoscale runs with pixel of about 250 mmicroscale runs with pixel of about 20 m

UPSCALINGExtension to other Countries?

PERSPECTIVES

IPHEN is a network among people belonging to universities, research institutions and services that for a lot of aims are interested in phenological monitoring and modelling. I wish to thank all people that co-operated to phenological monitoring and more specifically...

AKNOWLEDGMENTS

AKNOWLEDGMENTS - Grapevine

AKNOWLEDGMENTS – Black elder

AKNOWLEDGMENTS - Cypress

AKNOWLEDGMENTS – Olive tree

AKNOWLEDGMENTS – Locust tree

Thank you for your kind attention

IPHEN ANALYSIS MODEL – 1PRODUCTION OF THE FIRST GUESS MAPS

METEOROLOGICAL DRIVING VARIABLES - daily maximum and minimum temperatures (Tx and Tn) from 87 stations of Italian Agrometeorological Network of CRA – Cma.

DATA SPATIALIZATION - production of daily italian maps of Tx, Tn by means of the Shepard's method applied on data previously homogenized for height and aspect

PRODUCTION OF HOURLY TEMPERATURE MAPS (Parton – Logan model)

DATA PROCESSING WITH NHH MODELS with

- production of NHH maps cumulated from 1 January (1 october for cypress) until the analysis date

- production of first guess maps of the present phenological phase

MATCH BETWEEN OBSERVED AND SIMULATED DATA Error→

ERROR SPATIALIZATION with production of error maps.

CORRECTION OF FIRST GUESS MAPS with production of final maps

IPHEN ANALYSIS MODEL – 2PRODUCTION OF FINAL MAPS

Flowchart of Iphen model

IPHEN FORECAST MODEL

METEOROLOGICAL DRIVING VARIABLES

farecast data from today until the end of the “phenological season” coming from:

Data coming from GFS NWP model (1-10 day)

Decadal climatic normals from 10th day until the end of the season

NHH CUMULUS FROM 1TH JANUARY UNTIL TODAY: obtained from final maps applying the inverse equations of the NHH models (from phenological phase to cumulated NHH)

NHH CUMULUS FROM TODAY UNTIL THE FORECASTED PHENOLOGICAL PHASE: obtained applying NHH model to forecast temperatures.