Animal disease and plant pest dynamics and modelling. Animal disease and plant pest dynamics and modelling

Transcription

1 The Gambia. Model for G. morsitans 25 Animal disease and plant pest dynamics and modelling 2 Animal disease and plant pest dynamics and modelling Populations David Rogers Oxford University, UK david.rogers@zoo.ox.ac.uk Observ ed Model Puparia Model Tenerals Model Adults Month Data courtesy of ITC, The Gambia Climate-related trans-boundary pests. Rome February 28

2 The EDEN project Emerging Diseases in a changing European environment.preparing for an uncertain future European Union Framework 6 project, Climate-related trans-boundary pests. Rome February 28

3 EDEN Network Europe 48 Partners 24 countries

4 EDEN Structure Vertical Sub-Projects (SP) and Horizontal Integration Teams (HITs ( HITs) Disease Modelling West Nile Virus Tick borne diseases Rodent borne diseases HR Environmental Change Malaria Leishmaniasis African Platform LR Spatial Modelling Biodiversity Data Management

5 EDEN Aims 1) To describe the past 2) To explain the present 3) To predict the future

6 The Problem Three stages of invasion: 1.Arrival unpredictable chance event, but loaded by human activities. 2.Establishment depends on local climatic and other conditions. 3.Spread natural diffusion, trade,travel,, etc.

7 EDEN Modelling Aims 1) To develop statistical Risk Maps of EDEN s vectors and diseases. 2) To develop process-based biological models of pathogen, vector and host interactions.

8 What the User Wants a good model

9 Biological Models are put together as a series of components Wrong model assumption, or omission The model can never work the way it was intended

10 Statistical Models.. are put together as a series of (initially crude) correlations The model always has some descriptive power

11 Monthly variation in Land Surface Temperature (LST) and Sea Surface Temperature (SST) Black, blue, green = lower temperatures. Yellow, red white = higher temperatures

12 New Satellite imagery Terra/MODIS

13 Statistical Modelling

14 Towards disease burden mapping?! s l o o t y d a e Rift Valley Fever r Dengue e Visceral leishmaniasis u Cutaneous Yellow Fever leishmaniasis g n CrimeanCrimean -Congo Haemorrhagic Fever o t e u l B Probability of suitability

15 A good Early Warning System will have to be based on a biological rather than statistical description of diseases. From Statistics.... to Biology Vectors are often the most variable element of vector- borne diseases

16 Biological Modelling

17 In the beginning R = mbca µ 2 r e µ T R = basic reproductive number of the disease m = vector/host ratio b,c = transmission coefficients a = human biting rate µ = daily mortality rate T = extrinsic incubation period (days) r = rate of recovery of from infection all of these may be affected by climate/weather

18 Assumption of the simple Ro equation Things do not change in space e or through time

19 2 R = µ T R = 2 µ T 2 µ T 2 µ T 2 µ T 2 µ T EDEN R = R = R = R = R Disease Modelling 2 R = R = µ R = µ µ T 2 µ T 2 µ T 2 µ T 2 µ T 2 µ T making the mbca models e mbca more e realistic R = R = R = R = R = R mbca 2 e T R = 2 µ T 2 µ T 2 µ T 2 µ T 2 µ T 2 µ T mbca = R = R = R = R = R r 2 µ T 2 µ T 2 µ T 2 T 2 µ T 2 µ T R = R = R = R = R = R r µ 2 R = µ T R = 2 µ T 2 R = µ T 2 R = µ T 2 R = µ T 2 R = µ T R 2 R = µ T R = 2 µ T 2 R = µ T 2 R = µ T 2 R = µ T 2 R = µ T R

20 What can we learn from The Present?

21 The diseases we already have are changing naturally through space and time..usually we do not know the reasons why (trade, travel, global change?)

22 Aedes albopictus the Asian Tiger Mosquito Trade

23 Aedes albopictus original home (red), present distribution (+ orange), and port interceptions (yellow) Climatic dendrogram of global shipping ports Tatem, A.J., Hay, S.I. & Rogers, D.J. (26) PNAS 13,

24 Anopheles gambiae the malaria mosquito TraVeL

25 Number of passengers UK to sub-saharan Africa (upper) and Imported malaria cases, UK (lower) 35 No of passengers Total SSA East Africa West Africa Central Africa Southern Africa Number of UK travellers to SSA, Year i.e. approx 1 in 2, UK travellers to SSA return with P. falciparum malaria

26 Travels of the domestic chicken Google Earth, showing the locations of H5N1 records in Asia

27 H5N1 Risk maps With only environmental data. or also with poultry density Probability of suitability

28 Ixodes ricinus vector of Lyme disease and TBE EnvironmentaL Change

29 Lyme borreliosis and deer abundance in the UK Roe deer 3 LB cases acquired in England & 25 ρ = Roe deer population index Wales BBS UK data Coincidence or causal? Analysis by Dr Jörn Scharlemann, Oxford

30 New or newly emerging diseases do not necessarily behave in the same way in new hosts, new vectors or new continents..we rarely know the reasons why

31 Passing the baton there are also epidemiological Relay Races

32 Bluetongue Transmission Cycle Domestic sheep BT virus Culicoides spp. NEW EPIDEMIOLOGIES

33 Bluetongue. The Problem Nine introductions of different BTv strains into N. Africa and Europe since 1998

34 Bluetongue restriction zones October 27

35 Bluetongue. The Problem Three stages of invasion: 1.Arrival 2.Establishment Arrival wind field studies Establishment environmental studies/ modelling Spread wind field/environmental/modelling 3.Spread

36 Bluetongue. Establishment Predicted Risk maps for BT in Europe 26 Single predictor variable 1 predictor variables BT over-predicted? BT under-predicted?

37 Bluetongue. Establishment The UK situation BT confirmed case Probability of suitability

38 Bluetongue. Why 26? MODIS day-time temperature profiles in Europe

39 Culex pipiens vector of West Nile Virus (WNV) NEW EPIDEMIOLOGIES

40 The Spread (and numbers) of Human West Nile Virus cases in the USA 1999 to Human cases are the darker colors

41 West Nile Virus Transmission Cycle Mosquito vector Incidental infections West Nile virus Bird reservoir hosts Incidental infections At least 7 N. American mosquito species have been implicated in WNV transmission

42 How can we prepare for The Future?

43 The diseases we already have are changing naturally through space and time..usually we do not know the reasons why (trade, travel, global change?)..but some may be due to global warming

44 It is uncertain that global warming will cause an increase in all vector-borne diseases For example tick-borne encephalitis and rodent-borne hantaviruses appear to do better at lower temperatures and for different reasons.

45 Changes in recorded Minimum Temperatures across Europe between the 198 s and 199 s. Units are.1 degrees Celsius; increases in yellow/red, decreases s in blue

46 We do these things not because they are easy, but because they are hard. President J.F. Kennedy, on committing the USA to landing a man on the moon before the end of the 196s.

47 Network of independent but synergistic biological and non-biological factors Examples of data from Estonia, Latvia, Lithuania. National herd (1s) sheep & goats Agricultural 2 Change 197 Estonia cattle Decline of agriculture Area (1 ha) Estoni a 1975 Socio-economic transition? Sociological Change wooded agricultural pasture Regeneration of shrubs Industrial employees (1s) pollution emissions Higher unemployment More wealth & leisure Human behaviour Greater human exposure to ticks in forests? Environmental Change Increase in rodent populations? (transmission hosts) Data and analysis EDEN TBD Lithuania TBE cases TBE cases TBE cases Economic changes Estonia Latvia Lithuania Reduced industrial pollution Higher Increased TBE incidence Disease Incidence Mean nymphs, May & June Latvia Vector More infected Abundance ticks Sumilo et al, PLoS ONE 27, 6, e5 25 Mean daily max temp Roe deer/1 ha More ticks Wild boar/1 ha 2 More hosts for adult ticks 6 4 Host abundance Environmental awareness? Global brightening?? Climate Change Sudden increase in Spring temperature 1.Mar-2.Apr Estoni a Increased co-feeding transmission of TBEV Estonia, Latvia Mean daily max temp Apr Vector Behaviour

48 What we don t know is really what we need, And what we know is of no use whatever! Faust laments being unable to halt the plague Faust, by Rembrandt

49 Conclusions Global Warming and Animal Diseases. 1.Learn from the past. 2.Understand the present. Background knowledge, e.g. of pathogen, vector and host distributions, and how these are changing through space and time (environmental monitoring/satellite data/sentinel animals) 3.Explore understanding in a modelling framework 4.Develop forecasting tools - Disease Monitoring and Early Warning Systems (DMEWS) 5.Do not forget the public! Public/Veterinary Health and Information systems (to recruit, advise, forewarn and protect) 6.Predict the future, but expect the unexpected!

50 Thank you

51

52

53 Problems with Inter-Disciplinary studies CLIMATE SALES Epidemiologist Climatologist

54 Statistical Modelling: the long and winding road models are scary! Birth Death Biological or process based Emigration models are really Immigration? hard! Transmission Infection Recovery Immunity A Model! (The R equation) mbca 2e µt R = µr Snoopy Ph.D. A Risk Map!

55 Satellite predicted presence/absence maps for 4 WNV mosquito vectors. Darsie & Ward map (in yellow) At least 7 N. American mosquito species have been implicated in WNV transmission TALA Research, Oxford