statistical methods for improving authentication …...maintenance of databank wine data arbitration...
TRANSCRIPT
![Page 1: STATISTICAL METHODS FOR IMPROVING AUTHENTICATION …...Maintenance of databank Wine data Arbitration of disputes Analysis of samples Development and validation of methods Validation](https://reader035.vdocuments.us/reader035/viewer/2022071212/602720b19dbd8f6e1e34588e/html5/thumbnails/1.jpg)
Climatic and geographical dependence of H, C and O stable
isotope ratios of Italian wine
N. Dordevic, F. Camin, R. Wehrens, M. Neteler, G. J. Postma, L. M. C. Buydens,
Nikola Dordevic
![Page 2: STATISTICAL METHODS FOR IMPROVING AUTHENTICATION …...Maintenance of databank Wine data Arbitration of disputes Analysis of samples Development and validation of methods Validation](https://reader035.vdocuments.us/reader035/viewer/2022071212/602720b19dbd8f6e1e34588e/html5/thumbnails/2.jpg)
Stable isotope ratios of wine
(D/H)1
(D/H)2
δ13C
δ18O
![Page 3: STATISTICAL METHODS FOR IMPROVING AUTHENTICATION …...Maintenance of databank Wine data Arbitration of disputes Analysis of samples Development and validation of methods Validation](https://reader035.vdocuments.us/reader035/viewer/2022071212/602720b19dbd8f6e1e34588e/html5/thumbnails/3.jpg)
France
400
Germany
200
Italy
400
Spain
200 U. Kingdom
4
Austria
50
Greece
50
Luxembourg
4
Portugal
50
Wine data Maintenance of databank
Arbitration of disputes
Analysis of samples
Development and
validation of methods
Validation of data
Training
Malta
4
Slovakia
15
Slovenia
20
Hungary
50 Czech Rep.
20
Cyprus
10
Bulgaria
30
Romania
70
Wine Databank EC Reg. Nº 555/2008
![Page 4: STATISTICAL METHODS FOR IMPROVING AUTHENTICATION …...Maintenance of databank Wine data Arbitration of disputes Analysis of samples Development and validation of methods Validation](https://reader035.vdocuments.us/reader035/viewer/2022071212/602720b19dbd8f6e1e34588e/html5/thumbnails/4.jpg)
Aim of the study
1. To evaluate relationships between wine (D/H)1,
(D/H)2, δ13C and δ18O and climatic and
geographic parameters of provenance areas.
2. To build a model able to explain relationships
between wine isotope ratios and climatic and
geographic parameters of provenance areas.
![Page 5: STATISTICAL METHODS FOR IMPROVING AUTHENTICATION …...Maintenance of databank Wine data Arbitration of disputes Analysis of samples Development and validation of methods Validation](https://reader035.vdocuments.us/reader035/viewer/2022071212/602720b19dbd8f6e1e34588e/html5/thumbnails/5.jpg)
Materials and methods Variables Data
type Resolution
Source
Date of harvest static day Wine DB
Latitude static point www.findlatitudeandlogitude.com
Longitude static point www.findlatitudeandlogitude.com
Elevation static 20m Italian elevation model
Distance from the sea
static 250m Derived from elevation map in GIS
Amount of precipitation [mm/day]
dynamic 25km ECA&D, http://www.ecad.eu
Maximum daily temperature
dynamic 25km ECA&D
Minimum daily temperature
dynamic 25km ECA&D
Mean daily temperature
dynamic 25km ECA&D
δ18O of precipitation
static 37km Bowen et al. (2005)
δ2H of precipitation
static 37km Bowen et al. (2005)
• Official samples from the
Italian Wine Databank
from 2000 to 2010 are
considered.
• Explorative analyses and
linear modelling
![Page 6: STATISTICAL METHODS FOR IMPROVING AUTHENTICATION …...Maintenance of databank Wine data Arbitration of disputes Analysis of samples Development and validation of methods Validation](https://reader035.vdocuments.us/reader035/viewer/2022071212/602720b19dbd8f6e1e34588e/html5/thumbnails/6.jpg)
Results (D/H)1
99.33
99.74
100.18
100.52
100.76
101.09
101.3
101.43
101.62
101.67
101.83
102.06
102.07
102.45
102.91
103.12
103.3
103.42
104.12
![Page 7: STATISTICAL METHODS FOR IMPROVING AUTHENTICATION …...Maintenance of databank Wine data Arbitration of disputes Analysis of samples Development and validation of methods Validation](https://reader035.vdocuments.us/reader035/viewer/2022071212/602720b19dbd8f6e1e34588e/html5/thumbnails/7.jpg)
Results : explorative analyses
(D/H)1
(D/H)2
C13
O18
harvest date
latitude
longitude
elevation
distance
O18 MW
D/H MW
mean T
min T
max T
precipitation
(D/H
)1
(D/H
)2C13
O18
harvest
date
latitude
long
itude
elev
ation
distanc
e
O18
MW
D/H
MW
mea
n T
min T
max
T
prec
ipita
tion
-1.0
-0.5
0.0
0.5
1.0
![Page 8: STATISTICAL METHODS FOR IMPROVING AUTHENTICATION …...Maintenance of databank Wine data Arbitration of disputes Analysis of samples Development and validation of methods Validation](https://reader035.vdocuments.us/reader035/viewer/2022071212/602720b19dbd8f6e1e34588e/html5/thumbnails/8.jpg)
Results : explorative analyses
-0.06 -0.04 -0.02 0.00 0.02 0.04 0.06
-0.0
6-0
.04
-0.0
20
.00
0.0
20
.04
0.0
6
PC1 (40.1%)
PC
2 (
18
.6%
)
-40 -20 0 20 40
-40
-20
02
04
0
(D/H)1(D/H)2C13
O18
harvest date
latitude
longitude elevation
distance
O18 MW(D/H) MW
meanT
min T
max T
precipitation
![Page 9: STATISTICAL METHODS FOR IMPROVING AUTHENTICATION …...Maintenance of databank Wine data Arbitration of disputes Analysis of samples Development and validation of methods Validation](https://reader035.vdocuments.us/reader035/viewer/2022071212/602720b19dbd8f6e1e34588e/html5/thumbnails/9.jpg)
Results: linear modelling
![Page 10: STATISTICAL METHODS FOR IMPROVING AUTHENTICATION …...Maintenance of databank Wine data Arbitration of disputes Analysis of samples Development and validation of methods Validation](https://reader035.vdocuments.us/reader035/viewer/2022071212/602720b19dbd8f6e1e34588e/html5/thumbnails/10.jpg)
Results: linear modelling Regression coefficients for the selected variables
Coefficient_size
Date of harvest
Distance to the sea
Elevation
Latitude
Longitude
Precipitation
Temperature
-0.4 -0.2 0.0 0.2
C13 O18
Date of harvest
Distance to the sea
Elevation
Latitude
Longitude
Precipitation
Temperature
DH1
-0.4 -0.2 0.0 0.2
DH2R2 = 0.42
R2 = 0.71 R2 = 0.24
R2 = 0.30
![Page 11: STATISTICAL METHODS FOR IMPROVING AUTHENTICATION …...Maintenance of databank Wine data Arbitration of disputes Analysis of samples Development and validation of methods Validation](https://reader035.vdocuments.us/reader035/viewer/2022071212/602720b19dbd8f6e1e34588e/html5/thumbnails/11.jpg)
Conclusions
1. δ18O and (D/H)1 have the strongest relationship with
climate and location.
2. The dominant variables are latitude, δ18O and δ2H
of MW and temperature.
3. Models may be used in wine authenticity
assessments.
![Page 12: STATISTICAL METHODS FOR IMPROVING AUTHENTICATION …...Maintenance of databank Wine data Arbitration of disputes Analysis of samples Development and validation of methods Validation](https://reader035.vdocuments.us/reader035/viewer/2022071212/602720b19dbd8f6e1e34588e/html5/thumbnails/12.jpg)
• F. Camin, N. Dordevic, R. Wehrens, M. Neteler, L. Delucchi, G.
Postma, L. Buydens (2014) Climatic and geographical
dependence of the H, C and O stable isotope ratios of Italian wine. Analytica Chimica Acta (accepted, in press).
![Page 13: STATISTICAL METHODS FOR IMPROVING AUTHENTICATION …...Maintenance of databank Wine data Arbitration of disputes Analysis of samples Development and validation of methods Validation](https://reader035.vdocuments.us/reader035/viewer/2022071212/602720b19dbd8f6e1e34588e/html5/thumbnails/13.jpg)
Acknowledgment
Anti-Fraud Department of Italian Ministry of
Agricultural, Food and Forestry Policy, owner of the
Italian wine databank