arthur rylah institute management of on-farm risk to livestock from bovine tb in white-tailed deer...

Arthur RylahInstitute

Flora, Fauna &Freshwater Research

Arthur Rylah Institute

Management of on-farm risk to livestock from bovine TB in white-tailed deer within Deer Management Unit 452:

Predictions from a spatially-explicit model

David Ramsey, Daniel O’Brien, Rick Smith, Melinda Cosgrove, James Averill, Stephen Schmitt, Brent Rudolph




Spatial model of bTB in WTD in DMU452

• Recent development of a spatial model of TB in WTD has examined the efficacy of management options for DMU452 [Ramsey et al. 2014. J. Wildl. Manage 78(2):240-254.]

• Modelled scenarios included • Increase in harvest• Vaccination• Increase in harvest + vaccination• The effect of baiting

• All scenarios were examined as to their efficacy to eradication of TB from WTD within 30 years




Findings from the deer TB model

• Current MDNR management is unlikely to eradicate TB over the next three decades

• Eradication is possible within three decades, but is likely to require substantial increases in current harvest and/or vaccination

• TB establishment in a previously TB-free region is ~8 times more likely if baiting occurs during the hunting season

• In the meantime, cattle on farms within DMU 452 continue to be at-risk of TB infection from WTD




The way forward?

• If complete eradication of TB from WTD is too difficult, should focus change to risk mitigation for livestock?

• Acceptable management options may exist that will minimize risk of on-farm transmission from WTD to livestock

• Extend current spatial model to include transmission of TB from WTD to livestock




Modelling livestock transmission

• A spatial “livestock” layer was created for the existing model using records from the Michigan Department of Agriculture & Rural Development– Farm location– Area of cleared pasture– Stocking rate

• Data on the TB cattle herd infection (aka breakdown) rate 2003 – 2012 was also collated and used to calibrate transmission

• TB transmission dependent on stocking rate and contact rate with infected WTD




DMU 452, farm locations & their relative areas




Mean cattle herd infection rate/year vs predicted

Me

an

bre

akd

ow

ns/

yea

r

Obs Predicted

05

10

15




On-farm risk by location (no mitigation)

High risk

Mod risk

Low risk




Effect of management on transmission to livestock

• Evaluate effects of various management options on the risk of transmission to livestock (herd infections/breakdowns)

• Management of WTD within DMU452– Increasing harvest rate– Vaccination– Increase harvest + vaccination

• On-farm management practices– Restricting contact between WTD and cattle on farms – Local WTD control in the vicinity of farms

• Scenarios examined with and without baiting




Effect of management of WTD in DMU452 on cattle herd infection rates on farms




Increasing harvest rates

current

current

1.25x

1.5x

2x

1.25x

1.5x

2x

2x

3x

Multiple of current harvest

Antlered Antlerless

02

04

06

08

01

00

1 2 3 4 1 2 3 4 Scenarios




Effects of increasing harvest on HB (with baiting)

years

bre

akd

ow

ns

1 3 5 7 9 11 14 17 20 23 26 29

0

1

2

3

4

1 per year

1 per 5 years

years

Pro

ba

bili

ty o

f ze

ro b

rea

kdo

wn

s

1 3 5 7 9 11 14 17 20 23 26 29

0.0

0.2

0.4

0.6

0.8

1.01.5x 1.5x1.25x 1.25x

2.0x 2.0x2.0x 3.0x

Antler Antlerless




Effects of increasing harvest on HB (no baiting)

years

bre

akd

ow

ns

1 3 5 7 9 11 14 17 20 23 26 29

0

1

2

3

4

1 per year

1 per 5 years

years

Pro

ba

bili

ty o

f ze

ro b

rea

kdo

wn

s

1 3 5 7 9 11 14 17 20 23 26 29

0.0

0.2

0.4

0.6

0.8

1.01.5x 1.5x1.25x 1.25x

2.0x 2.0x2.0x 3.0x

Antler Antlerless




Harvest + 90% vaccinated annually (with baiting)

years

bre

akd

ow

ns

1 3 5 7 9 11 14 17 20 23 26 29

0

1

2

3

4

1 per year

1 per 5 years

years

Pro

ba

bili

ty o

f ze

ro b

rea

kdo

wn

s

1 3 5 7 9 11 14 17 20 23 26 29

0.0

0.2

0.4

0.6

0.8

1.01.5x 1.5x1.25x 1.25x

2.0x 2.0x2.0x 3.0x

Antler Antlerless




Harvest + 90% vaccinated annually (no baiting)

years

bre

akd

ow

ns

1 3 5 7 9 11 14 17 20 23 26 29

0

1

2

3

4

1 per year

1 per 5 years

years

Pro

ba

bili

ty o

f ze

ro b

rea

kdo

wn

s

1 3 5 7 9 11 14 17 20 23 26 29

0.0

0.2

0.4

0.6

0.8

1.01.5x 1.5x1.25x 1.25x

2.0x 2.0x2.0x 3.0x

Antler Antlerless




Vaccination only (with baiting)

years

bre

akd

ow

ns

1 3 5 7 9 11 14 17 20 23 26 29

0

1

2

3

4

1 per year

1 per 5 years

years

Pro

ba

bili

ty o

f ze

ro b

rea

kdo

wn

s

1 3 5 7 9 11 14 17 20 23 26 29

0.0

0.2

0.4

0.6

0.8

1.0

50% coverage90% covereage




Vaccination only (no baiting)

years

bre

akd

ow

ns

1 3 5 7 9 11 14 17 20 23 26 29

0

1

2

3

4

1 per year

1 per 5 years

years

Pro

ba

bili

ty o

f ze

ro b

rea

kdo

wn

s

1 3 5 7 9 11 14 17 20 23 26 29

0.0

0.2

0.4

0.6

0.8

1.0





Effect of on-farm management




Restriction of contact between WTD and cattle

• Baseline model assumes unrestricted contact between WTD and cattle on farms

• Examined the effect of restricting contact on cattle herd infections

• Practically this can be achieved (for example) by– Improved fencing– Restricting access to food sources




On-farm contact reduction (%)

No reduction

High risk

Mod risk

Low risk





20% reduction

High risk

Mod risk

Low risk





50% reduction

High risk

Mod risk

Low risk





80% reduction

High risk

Mod risk

Low risk





90% reduction

High risk

Mod risk

Low risk




Local control of WTD

• Manage WTD in the vicinity of farms only• Less expensive option than management of

deer across the entire DMU• What size buffer would be adequate to

achieve significant reduction in herd infections?




5 km buffer around farms (32% of total DMU area)




Local vaccination within 5km buffer (no baiting)

years

bre

akd

ow

ns

1 3 5 7 9 11 14 17 20 23 26 29

0

1

2

3

4

1 per year

1 per 5 years

years

Pro

ba

bili

ty o

f ze

ro b

rea

kdo

wn

s

1 3 5 7 9 11 14 17 20 23 26 29

0.0

0.2

0.4

0.6

0.8

1.0





Local control within 5km buffer (no baiting)

years

bre

akd

ow

ns

1 3 5 7 9 11 14 17 20 23 26 29

0

1

2

3

4

1 per year

1 per 5 years

years

Pro

ba

bili

ty o

f ze

ro b

rea

kdo

wn

s

1 3 5 7 9 11 14 17 20 23 26 29

0.0

0.2

0.4

0.6

0.8

1.0

20% cull50% cull




An aside: If prevalence in deer has gone down, why are cattle herds still getting infected?

• Speculation that because cattle herds are still becoming infected, there must be some wild species (other than deer) infecting cattle

• At least two existing lines of evidence argue against such speculation– Modelling– Spatial variations in deer prevalence




Figure 3A, Ramsey et al., 2014, J. Wildl. Mgt. 78(2):245.


1.5x 1.5x1.25x 1.25x

2.0x 2.0x2.0x 3.0x

Antler Antlerless

~4 years ~4 years

Deer prevalence:~0.7%






1.5x 1.5x1.25x 1.25x

2.0x 2.0x2.0x 3.0x

Antler Antlerless

~8 years ~8 years







1.5x 1.5x1.25x 1.25x

2.0x 2.0x2.0x 3.0x

Antler Antlerless

30 years







1.5x 1.5x1.25x 1.25x

2.0x 2.0x2.0x 3.0x

Antler Antlerless

~18 years ~18 years





TB prevalence in deer varies locally

200329N 6E: 3.0%DMU452: 1.7%





200329N 6E: 3.0%DMU452: 1.7%

200631N 5E: 3.1%DMU452: 2.3%

200630N 5E: 4.9%DMU452: 2.3%





200329N 6E: 3.0%DMU452: 1.7%

200631N 5E: 3.1%DMU452: 2.3%

200630N 5E: 4.9%DMU452: 2.3%

201227N 5E: 5.6%DMU452: 1.7%




The future: predicting local variations in TB prevalence




Conclusions (DMU wide control)

• Compared with TB control in WTD directly, management aimed at reduction of cattle herd infections requires much less effort ($)

• But… management needs to continue in perpetuity as TB remains in the wider deer population– Gains will be rapidly lost once management ceases (e.g. lifting

of baiting bans)

• A 25% increase in harvest and no baiting would halve the rate of cattle herd infections within 3-5 years and reduce it by 95% within 15 years

• Vaccination each year achieving 50% coverage would also achieve the same result




Conclusions (farm level control)

• Substantial reduction in the risk of herd infections is achieved if contact between WTD and cattle on farms is reduced by at least 80%

• Local control measures can also be effective

– Vaccinating at least 50% of WTD within 5 km of farms will reduce the cattle herd infection rate by 95% within 13 years

– Culling 50% of deer in addition to harvest within the 5 km buffer would reduce the herd infection rate by 95% within 10 years




Conclusions

• That cattle herd infections continue to occur annually despite reductions in TB in deer is expected, and consistent both with modeled predictions and what we already know about TB in deer

• Results suggest that without effective wildlife risk mitigation, TB prevalence in deer at the DMU452 scale will need to be maintained well below 1% before cattle herd infections do not consistently occur every year, and well below 0.1% before there is a high probability of having no annual cattle herd infections




Thank you

arthur rylah institute management of on-farm risk to livestock from bovine tb in white-tailed deer...

Documents

transmission of tb

farm transmission

spatial model of tb

bovine tb

baitingeffect of management

contact rate

atrisk of tb infection

risk of transmission