A dynamic model fordowny mildew primary
infection on grape
Bugiani R. Brunelli A. Collina M.
Rossi V. Giosué S. Caffi T. Girometta B.
Spanna F.
Many forecasting models forP.viticola primary infection,
were developed
None of them proved to be precise and robust
Current warning systems are mainly based on “Three 10 Rule”
In spite of the fact it is often unreliable
A new approach:- Pathosystem analysis- data and information collection- mathematic relationships build-up- dynamic simulation
ValidationEmilia-Romagna 1995 – 2002, several localitiesPiemonte 1999 –2002, several localitiesOltrepò Pavese 1998 – 2002
In 2003Emilia-RomagnaPiemonte
Oosporelatency Temperature
Water presence
Germination
Infection
zoosporangiasurvival
Zoosporesurvival
TemperatureRelative humidity
Zoosporeliberation
TemperatureLeaf wetness
Spread
Rainfall
Symptomsoccurrence
Incubation
TemperatureRelative Humidity
Water presence
THE
MODEL
THE
MODEL
Start
Oosporemature
Spore germination
Presenceof rain
No
Y
Presence of sporangia
Presence ofzoospore
End
Wetness
Presenceof water
Presenceof Rain
SurvivedSporangia
Presence ofwetness Infection
Zoospore death
Zoospores on the leaves
No
Y
No
No
No
No
NoY
Y
Y
Y
Y
1.Oospore latency
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.51/
1
15/1
29/1
12/2
26/2
12/3
26/3 9/4
23/4
IMO
1999
2000
2001
2002
IMO = f (T, VPD)
Index of Oospore Maturation(IMO)
THE
MODEL
1.Oospore latency
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.51/
1
15/1
29/1
12/2
26/2
12/3
26/3 9/4
23/4 7/5
21/5
IMOmin
IMOmax
Period of oospore
maturation
IMO
THE
MODEL
2. Oospore germination
Coorts of matureoospores
Frequency of mature oospores
Start of germination
f (R)
IMOmaxIMOmin
THE
MODEL Period of
oosporematuration
2. Oospore germination
10/4
15/4
20/4
25/4
30/4 5/5
10/5
15/5
20/5
25/5
30/5
1
0.8
0.6
0.4
0.2
0
IGO
IGO = f (T, VPD)Index of Oosporegermination (IGO)
THE
MODEL
3.Sporangia survival
10/4
15/4
20/4
25/4
30/4 5/5
10/5
15/5
20/5
25/5
30/5
SURmax= f (T, RH)
THE
MODEL
4.Sporangia germination5.Zoospore survival
f (T, LW)10
/4
15/4
20/4
25/4
30/4 5/5
10/5
15/5
20/5
25/5
30/5
f (LW)
THE
MODEL
6.Zoospores dispertion
f (R)10
/4
15/4
20/4
25/4
30/4 5/5
10/5
15/5
20/5
25/5
30/5
THE
MODEL
7.Infection
f (T, LW)10
/4
15/4
20/4
25/4
30/4 5/5
10/5
15/5
20/5
25/5
30/5
THE
MODEL
8.Incubation and symptom occurrence
f (T, RH)10
/4
15/4
20/4
25/4
30/4 5/5
10/5
15/5
20/5
25/5
30/5
THE
MODEL
Date of oospore observationGermination observedGermination expected
Asti 2000
1
1
2
2
3
3
4
4
5
5
6
6
30/3 4/4
9/4
14/4
19/4
24/4
29/4 4/5
9/5
14/5
19/5
24/5
29/5
y = 1.07x - 3.3R2 = 0.94
0
10
20
30
40
50
60
0 10 20 30 40 50 60Observed (days from 30/3)
Stim
ati
Asti 1999-2003
THE
VALI
DATI
ONS
–Oos
pore
germ
inat
ion
Velocity of oosporegermination
Estimated values of the modelvs
Observed values for oosporesoverwintered in vineyard(floating disk method)
Asti 1999-2003
Date primaty symptoms occurrenceTH
E VA
LIDATI
ONS
–Sy
mpt
omoc
curr
ence
Model’s estimated date vs Observed date in vineyard1/
4
8/4
15/4
22/4
29/4 6/5
13/5
20/5
27/5 3/6
10/6
17/6
24/6 1/7
8/7
15/7
22/7
29/7
0
10
20
30
40
50
60
70R (mm)
Oosporegermination
Zoosporerelease
Spread of zoospores
Infection
End of incubation
Occurrence
Symptomsoccurrence
05
10152025303540
%
28-1
/4
2-6/
4
7-11
/4
12-1
6/4
17-2
1/4
22-2
6/4
27-1
/5
2-6/
5
12-1
6/5
17-2
1/5
22-2
6/5
27-3
1/5
7-11
/5
1-5/
6
6-10
/6
11-1
5/6
16-2
0/6
21-2
5/6
26-3
0/6
1-5/
7
35
05
1015202530 % Infection
28-1
/4
2-6/
4
7-11
/4
12-1
6/4
17-2
1/4
22-2
6/4
27-1
/5
2-6/
5
12-1
6/5
17-2
1/5
22-2
6/5
27-3
1/5
7-11
/5
1-5/
6
6-10
/6
11-1
5/6
16-2
0/6
21-2
5/6
26-3
0/6
1-5/
7
28-1
/4
2-6/
4
7-11
/4
12-1
6/4
17-2
1/4
22-2
6/4
27-1
/5
2-6/
5
12-1
6/5
17-2
1/5
22-2
6/5
27-3
1/5
7-11
/5
1-5/
6
6-10
/6
11-1
5/6
16-2
0/6
21-2
5/6
26-3
0/6
1-5/
705
10152025303540
% 10 cm shoots
38 cases(locality x year)
THE
VALI
DATI
ONS
–Sy
mpt
omoc
curr
ence
THE
VALI
DTZ
IONS
–Sy
mpt
oms
occu
rren
ce308 simulations
no yes
232 075.3% 0%
17 59
5.5% 19.2%yes
no
Infe
ctio
nex
pect
edInfection observed
291 correct17 uncorrect
False alarms
Correct predictionsTH
E VA
LIDATI
ONS
–Sy
mpt
om o
ccur
renc
e
Oltrepò PV 2001
1/4
8/4
15/4
22/4
29/4 6/5
13/5 1/4
8/4
15/4
22/4
29/4 6/5
13/5
20/5
27/5
Carpineta (MO) 1996 Panocchia (PR) 1997
1/4
8/4
15/4
22/4
29/4 6/5
13/5
20/5
27/5 3/6
10/6
17/6
0
10
20
30
40
50
60
70
3/6
20/5
Correct predictionsTH
E VA
LIDATI
ONS
–Sy
mpt
om o
ccur
renc
eAlba (CN) 2002
0
20
40
60
80
100
0
20
40
60
80
100% of affected leaves
20/5
23/5
26/5
29/5 1/6
4/6
7/6
10/6
13/6
16/6
19/6
20/5
23/5
26/5
29/5 1/6
4/6
7/6
10/6
13/6
16/6
19/6
20/5
23/5
26/5
29/5 1/6
4/6
7/6
10/6
13/6
16/6
19/6
Ger
min
atio
nO
ospo
res
2/53/56/5
18/5
20/521/525/525/5
Infe
ctio
n2-3/54/5
8-9/5
18-19/5
23-25/523-26/525-29/527-29/5
End
of
incu
batio
n
13-16/514-16/515-18/5
24-27/5
29/5-1/629/5-1/631/5-3/6
1-4/6
12/415/419/4
7/5
10/512/516/518/5
Star
t of
Ger
min
atio
n
sporulations
THE
VALI
DATI
ONS
–Sy
mpt
om o
ccur
renc
eFalse alarms
28-1
/4
2-6/
4
7-11
/4
12-1
6/4
17-2
1/4
22-2
6/4
27-1
/5
2-6/
5
12-1
6/5
17-2
1/5
22-2
6/5
27-3
1/5
7-11
/5
1-5/
6
6-10
/6
11-1
5/6
16-2
0/6
21-2
5/6
26-3
0/6
1-5/
7
28-1
/4
2-6/
4
7-11
/4
12-1
6/4
17-2
1/4
22-2
6/4
27-1
/5
2-6/
5
12-1
6/5
17-2
1/5
22-2
6/5
27-3
1/5
7-11
/5
1-5/
6
6-10
/6
11-1
5/6
16-2
0/6
21-2
5/6
26-3
0/6
1-5/
705
10152025303540
% 10 cm shoots
01234567
N.
THE
VALI
DATI
ONS
–Sy
mpt
om o
ccur
renc
eFalse alarms
Lavezzola (RA) 1998
1/4
8/4
15/4
22/4
29/4 6/5
13/5
20/5
27/5 3/6
10/6
0
10
20
30
40
50
60
70S.Agata sul Santerno (RA) 1997
17/61/4
8/4
15/4
22/4
29/4 6/5
13/5
20/5
27/5 3/6
10/6
Castel Boglione
Asti
1/4
8/4
15/4
22/4
29/4 6/5
13/5
20/5
27/5 3/6
10/6
17/6
24/6 1/7
8/7
15/7
22/7
29/7
0
10
20
30
40
50
60
70Serralunga
THE
VALI
DATI
ONS
–Ye
ar20
03
0
10
20
30
40
50
60
70
Piacenza
Castelfranco Emilia1/
48/
415
/422
/429
/4 6/5
13/5
20/5
27/5 3/6
10/6
17/6
24/6 1/7
8/7
15/7
22/7
29/7
Altedo
THE
VALI
DATI
ONS
–Ye
ar20
03
The forecasting model proved to be: detailedIt followed, step by step, the whole infection processaccurateIt corretly estimated 94% of the cases It never provided wrong negative prognosisrobustIt estimate either early and late infectionsIt adapt itself with several epidemiological conditions
The model gave some false alarms
Conc
lusion
s and therefore it can be furtherly improved