An epidemiological approach for the deployment of disease … · 2011. 8. 22. · An...
Transcript of An epidemiological approach for the deployment of disease … · 2011. 8. 22. · An...
An epidemiological approach for the deployment of disease control in successive crops
RMT modélisation - 29 sept. 2009, Paris
UMR BIO3P - équipe Epidémiologie Sol et Systèmes (EPSOS)
INRA – Centre de Rennes
Doug Bailey, Sylvain Poggi, Vincent Faloya
Objectives
– Devise models for disease and inoculum dynamics across successive crops.
– Extend the models to allow for inherent variability.
– Parameterise the models for chemical, biological and cultural control.
– Identify criteria for invasion and persistence.
– Use the models to optimise control.
RMT modélisation - 29 sept. 2009, Paris
Approach
Population dynamics
Epidemic processes
Macro
Meso
Micro
Spatio-temporalmodels
Temporalmodels
Pathozonedynamics
Modelling Experimentation
Field/Micro-plot
Micro-plot/Microcosm
MicrocosmEpidemic componentsGrowth rates
Primary infectionSecondary infection
Pathozonecomponents
Disease maps
RMT modélisation - 29 sept. 2009, Paris
Start point: Model derivation (single crop)
Susceptible
Exposed
Infected
Removed
( ) ( )d1
d p s
S Nb I N r X r I S
t κ= − − +
( )d
d p s i
Er X r I S rE
t= + −
d
d i r
IrE r I
t= −
d
d r
Rr I
t=
RMT modélisation - 29 sept. 2009, Paris
Important features of epidemic behaviour
Stochasticity(What is the risk of disease?)
Dynamically generated variability(Do small differences in control early in an epidemic lead to large difference in final
disease levels?)
Invasion thresholds(Are there critical combinations of parameters that lead to invasion or control?)
Persistence and hidden infestation(Does the quality of an epidemic affect persistence into the next susceptible crop?)(Does cryptic infection pose a risk for persistence of disease in future crops?)
RMT modélisation - 29 sept. 2009, Paris
Important features of epidemic behaviour
Stochasticity(What is the risk of disease?)
Dynamically generated variability(Do small differences in control early in an epidemic lead to large difference in final
disease levels?)
Invasion thresholds(Are there critical combinations of parameters that lead to invasion or control?)
Persistence and hidden infestation(Does the quality of an epidemic affect persistence into the next susceptible crop?)(Does cryptic infection pose a risk for persistence of disease in future crops?)
RMT modélisation - 29 sept. 2009, Paris
Stochasticity(What is the risk of disease?)
0.000 0.001 0.002 0.003 0.004 0.005
0.00
0.25
0.50
0.75
1.00
Like
liho
od
of
dis
ease
-fre
e si
tes
rp (rate of primary infection)
Disease riskDisease progress
RMT modélisation - 29 sept. 2009, Paris
Important features of epidemic behaviour
Stochasticity(What is the risk of disease?)
Dynamically generated variability(Do small differences in control early in an epidemic lead to large difference in
final disease levels?)
Invasion thresholds(Are there critical combinations of parameters that lead to invasion or control?)
Persistence and hidden infestation(Does the quality of an epidemic affect persistence into the next susceptible crop?)(Does cryptic infection pose a risk for persistence of disease in future crops?)
RMT modélisation - 29 sept. 2009, Paris
Dynamically generated variability(Does a small amount of control early in an epidemic lead to
large difference in final levels of disease?)
5 10 15 205 10 15 2001020304050
5 10 15 205 10 15 20
01020304050
Time (Days)
Nu
mb
er o
f d
ise
ase
d p
lan
ts
(a) Control of primary infection
(b) Control of secondary infection157
75
-Tv+Tv
RMT modélisation - 29 sept. 2009, Paris
0 500 1000 1500 2000 2500
0
20
40
60
80
100
Biofumigation of R. solani in sugar beet by B. napus(Can we exploit DGV for control of field epidemics?)
No mustard
Mustard removed
Mustard incorporated
Supervisors:
Philippe LucasFrançoise Montfort
Thierry Doré
Collaborator for epidemiogical analysis:
Joao Filipe
Natacha Motisi (PhD)
1ary
2ary
( )d
d p s
Sr X rI S
t=− +
( )d
d p s
Ir X rI S
t= +
d
d d
Xr X
t=−
2dexp(-0.5(log( / )/ ) )
d
ss s s
rt
tα γ β=
Susceptible
Infected
Inoculum
Secondary infection
% d
isea
se p
lant
s
Thermal time
RMT modélisation - 29 sept. 2009, Paris
0 500 1000 1500 2000 25000
1e-6
2e-6
3e-6
4e-6
0 500 1000 1500 2000 25000.00000
0.00002
0.00004
0.00006
0.00008
0.00010
Net rate of primary infection Net rate of secondary infection
No mustard
Mustard removed
Mustard incorporated
Biofumigation of R. solani in sugar beet by B. napus(Are the large differences in final levels of disease caused by
control of primary or secondary infection?)
RMT modélisation - 29 sept. 2009, Paris
Important features of epidemic behaviour
Stochasticity(What is the risk of disease?)
Dynamically generated variability(Do small differences in control early in an epidemic lead to large difference in final
disease levels?)
Invasion thresholds(Are there critical combinations of parameters that lead to invasion or
control?)
Persistence and hidden infestation(Does the quality of an epidemic affect persistence into the next susceptible crop?)(Does cryptic infection pose a risk for persistence of disease in future crops?)
RMT modélisation - 29 sept. 2009, Paris
Distance (r ) between donor and recipient sites
0 5 10 15 20
Pro
babi
lity
of c
olon
isat
ion
0.0
0.2
0.4
0.6
0.8
1.0
Pc
rc
a) b) c)
r > rc
r < rc
Donor Recipient
r
How do spatial invasion thresholds arise?
New Phyt. (2000) 146: 535-544
Fungal hyphae Pathozone Population
RMT modélisation - 29 sept. 2009, Paris
The effect of resource strength on the probability of colonisation between sites
0 2 4 6 8 10 12 14 16 18 20
Pro
babi
lity
of c
olon
isat
ion
(P )
0.0
0.2
0.4
0.6
0.8
1.0
Pc
Distance between donor and recipient
Low nutrientHigh nutrient
RMT modélisation - 29 sept. 2009, Paris
Experimental validation in microcosms(Can a soil-borne plant pathogen exploit an invasion threshold?)
Probability of colonisation (P) between donor and recipient.
0.0 0.2 0.4 0.6 0.8 1.0
Pro
port
ion
of p
atch
es s
how
ing
inva
sive
spr
ead
0.0
0.2
0.4
0.6
0.8
1.0
6 DAYS12 DAYS
NON INVASIVESPREAD
INVASIVE SPREAD
18DAYS
RMT modélisation - 29 sept. 2009, Paris
Controlling the invasive spread of disease(Can we use biological agents to block invasion?)
0.0
0.2
0.4
0.6
0.8
1.0
0 5 10 15 20 2545678910
Prob
abili
ty o
f di
seas
eDistance (mm)
0.0
0.2
0.4
0.6
0.8
1.0
0 5 10 15 20 2545678910
Prob
abili
ty o
f di
seas
e
Distance (mm)
Tim
e (d
ays)
-BIOCONTROL +BIOCONTROL
RMT modélisation - 29 sept. 2009, Paris
Proportion of sites removed from a population
0.0 0.2 0.4 0.6 0.8 1.0P
ropo
rtio
n of
pop
ulat
ions
sh
owin
g in
vasi
ve s
prea
d
0.0
0.2
0.4
0.6
0.8
1.0
Invasivespread
Non-invasivespread
42%
40%
Otten & Gilligan 2006:Eur J. Soil Sci: 57: 26-37Otten et al., 2004New Phyt: 163: 125-132
Proportion of sites removed from a population
0.0 0.2 0.4 0.6 0.8 1.0P
ropo
rtio
n of
pop
ulat
ions
sh
owin
g in
vasi
ve s
prea
d
0.0
0.2
0.4
0.6
0.8
1.0
Invasivespread
Non-invasivespread
Protection of individual sites(Do we need to protect the entire crop?)
RMT modélisation - 29 sept. 2009, Paris
Important features of epidemic behaviour
Stochasticity(What is the risk of disease?)
Dynamically generated variability(Do small differences in control early in an epidemic lead to large difference in final
disease levels?)
Invasion thresholds(Are there critical combinations of parameters that lead to invasion or control?)
Persistence and hidden infestation(Does the quality of an epidemic affect persistence into the next susceptible crop?)
(Does cryptic infection pose a risk for persistence of disease in future crops?)
RMT modélisation - 29 sept. 2009, Paris
2 4 6 8 10 12 14 16 18 20 22
0
10
20
30
40
50
2 4 6 8 10 12 14 16 18 20 22
0
10
20
30
40
50
Nu
mb
er o
f d
ise
ase
d p
lan
ts
Time (days)Time (days)
1st Season 2nd Season
0 5 10 15 20 25
0.0
0.5
1.0
1.5
2.0
2.5
Time (days)
Cfu
g-1
- Trichoderma
+ Trichoderma
Inter-season inoculum dynamics
Spread and control of disease in successive crops
RMT modélisation - 29 sept. 2009, Paris