Section). The fit of the statistical model to the global malaria was significantly better than the fit of two alternative, biological models to the same map, although this is not particularly surprising because the map was used in the construction of the statistical model, but not the biological one. We repeat, however, that the biological model has not been tested for its accuracy against any independent datasets. The model is a ssu med to be co rrect and therefore, it is further assumed, the map resulting from it must also be a correct picture of where malaria was distributed before human intervention. The statistically based model of global malaria distribution under present-day climate was then re-run with various climate scenarios for the future (Rogers and Randolph, 2000). Even under a relatively extreme scenario of climate change (the HadCM2 ‘High’ scenario: http://www.met-office.gov.uk/sec5/CR_dic/Brochure97/), there was remarkably little change in the predicted global distribution of malaria in the future, compared with the present day (Figures 2b and c). Unsurprisingly, areas predicted to be the most affected are those near the current edges of malaria’s global distribution; the southern United States, Turkey, Turkmenistan and Uzbekistan, Brazil and China. They also include some highland areas, for example, in East Africa, where malaria is predicted to appear for the first time, and some presently marginal areas that become too dry in the future and from which malaria is predicted to disappear (e.g. the eastern sea- board of India). In global terms, malaria is predicted to appear for the first time in areas in which about 360 million people live at the present time, and to disappear from areas where about 330 million people live at present (Figure 2c). The net difference (30 million peo- ple) is almost certainly not significant, given the uncertainties in the modelling. Zero net differences, however, hide the fact that almost 700 million people will be affected, one way or another, by even the modest changes predicted by the statistical model. 4.3. Malaria: Further Developments of Biological Models Using a different biological model for malaria in Africa, Thomas et al. came to conclusions rather more like those of the statistical D.J. ROGERS AND S.E. RANDOLPH358 Hungary Lithuania Sweden Lithuania Hungary Sweden Annual no. TBE cases 0 200 400 1960 65 70 75 80 85 90 95 2000 2005 Hungary 0 300 600 1960 65 70 75 80 85 90 95 2000 200 5 Lithuania 0 50 100 150 1960 65 70 75 80 85 90 95 2000 2005 Sweden 9 11 13 15 17 19 21 1960 65 70 75 80 85 90 95 2000 Year Mean temperature May-Aug 2 4 6 8 10 12 1960 65 70 75 80 85 90 95 2000 Mean temperature March-May Figure 3 Top: changes in the annual numbers of cases of tick-borne encephalitis in Sweden, Hungary and Lithuania, 1960–2004. The step in- creases in Sweden from 1983 to 1986 and again in 2000 are highlighted by horizontal lines showing mean levels in each period. Bottom: changes in mean spring (upper) and summer (lower) temperatures (taken from the in- terpolated climate surfaces prepared by the CRU, University of East Anglia) for 0.51 grid squares centred on Zala county Hungary, Siauliu in Lithuania and Stockholm in Sweden. Dotted horizontal lines show the 1960–2000 mean levels for each site. CLIMATE CHANGE AND VECTOR-BORNE DISEASES 367 Plate 8.4 Each virus of the tick-borne encephalitis complex occupies a distinct ‘eco-climatic’ space, illustrated here in bi-variate space defined by two of the most significant climatic variables that predict the distribution of each virus. NDVI (nor- malized difference vegetation index) is an indirect measure of moisture conditions. m, vector/host ratio a, biting rate µ, vector mortality rate T, extrinsic incubation period m a T µ temperature temperature temperature temperature ?? ? + - + +?- Plate 10.1 Likely effects of increasing temperature on the variables and param- eters of the R 0 equation. The net effect is indicated by the positive or negative symbol within each panel. Notice that a positive effect here might decrease transmission (e.g. the effect on m) or increase it (e.g. the effect on a). Probability = 00 - 0.349 = 0.50 - 0.549 = 0.45 - 0.499 = 0.65 - 1.0 = 0.35 - 0.449 = 0.55 - 0.649 = Observed = No prediction A Plate 10.2 (a) Global map of malaria distribution according to the WHO (1997) (yellow cross-hatching) and predicted distribution made from 1961–1990 global climate norms. Predictions were made using a discriminant analysis approach (Chapter 1 and (Rogers, 2000)) and are on a probability scale from zero (coloured red) to 1.0 (coloured green) (see inset legend) (model results: 78% correct with 14% false positives and 8% false negatives). (b) Predicted global distribution of malaria in 2050 under the HadCM2 High scenario of global warming. The model from (A) was run using these climate predictions to produce an estimate of malaria distribution in 2050 (colour scale as in A). The WHO map of malaria is shown for reference (yellow cross-hatching). (c) The difference between Plate 10.2A and B reveal the predicted changes in global malaria distribution in 2050. Areas coloured red are presently suitable for malaria but will become unsuitable (generally because of higher temperature or lower rainfall). Areas coloured green are presently unsuitable but are predicted to become suitable. All areas of no change (i.e. suitable or unsuitable, now and in 2050) are coloured white. Plates 10.2A–C from Rogers and Randolph (2000), with permission. Probability = 00 - 0.349 = 0.50 - 0.549 = 0.45 - 0.499 = 0.65 - 1.0 = 0.35 - 0.449 = 0.55 - 0.649 = Observed = No prediction B = presently suitable, becoming unsuitable by 2050 = presently unsuitable, becoming suitable by 2050 C Plate 10.2 (continued) Index Accessibility modeling, in global population distribution determination, 129–131 ADDS (Africa Data Dissemination Service), 67 ADEOS (Advanced Earth Observation Satellite) program, 63–64 Administrative boundary data, 68 A. duodenale, in STH infections, 221–224, 232–239, 243, 246–247 Ae. aegypti, 183, 185–197, 207, 209 global distribution of, 243 in dengue, 207, 307 in yellow fever, 210 Ae africanus, in yellow fever, 184–185 Ae. albopictus, 185 air travel risk routes of, 318, 330 global distribution of, 243 in yellow fever, 210 shipping risk routes of, 317 Ae. japonicus, 307–308 Africa east African highlands, increased incidence of malaria in, 370–374 infectious disease contagion in, 377 malaria in, 355, 370–377 Rift Valley Fever (RVF) in, 16, 29 WNV in, 307, 322–323 yellow fever and dengue fever in, 186–187, 194–195, 198, 200–209 AIC (Akaike Information Criterion), 23, 199 in yellow fever and dengue fever, 200–209 AIDS, 300–301 Airport malaria, 326–329 Albendazole, 234, 239, 247, 250 Altitudinal mask, in malaria transmission, 161–167 A. lumbricoides (roundworm), in STH infections, 222 An. atroparvus, 328 Ancillary data, in global population distribution determination, 124, 135–136, 148 An. gambiae, 307, 326–330 Anthrax and bioterrorism, 302–304 Asia, 298, 300 cholera in, 297–298 influenza in, 298–305 WNV in, 307, 322–323 yellow fever and dengue fever in, 186–187, 194–195, 198, 200–209 ASTER (Advanced Spaceborne Thermal Emission and Reflection) radiometer, in infectious disease distribution mapping, 55–56 AVHRR (Advanced Very High Resolution Radiometer) sensor, 37, 82, 198, see also NOAA and MODIS, 55–68 archives, 39 B. anthracis, 303 Biological maps, 3 Bioterrorism, 302 Black Death, due to bubonic plague, 295–297 Bootstrap sampling, 15–17, 30, 32 Brazil, yellow fever in, 184, 207–209 ADVANCES IN PARASITOLOGY VOL 62 ISSN: 0065-308X $35.00 DOI: 10.1016/S0065-308X(05)62013-1 Copyright r 2006 Elsevier Ltd. All rights of reproduction in any form reserved Contents of Volumes in This Series Volume 41 Drug Resistance in Malaria Parasites of Animals and Man . . . . . . . . . . . . 1 W. P ETERS Molecular Pathobiology and Antigenic Variation of Pneumocystis carinii 63 Y. N AKAMURA AND M. WADA Ascariasis in China. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109 P. W EIDONO,Z.XIANMIN AND D.W.T. CROMPTON The Generation and Expression of Immunity to Trichinella spiralis in Laboratory Rodents. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149 R.G. B ELL Population Biology of Parasitic Nematodes: Application of Genetic Markers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 219 T.J.C. A NDERSON, M.S. BLOUIN AND R.M. BRECH Schistosomiasis in Cattle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 285 J. D E BONT AND J. VERCRUYSSE Volume 42 The Southern Cone Initiative Against Chagas Disease . . . . . . . . . . . . . . . 1 C.J. S CHOFIELD AND J.C.P. DIAS Phytomonas and Other Trypanosomatid Parasites of Plants and Fruit. . . . . 31 E.P. C AMARGO Paragonimiasis and the Genus Paragonimus 113 D. B LAIR, Z B. XU AND T. AGATSUMA Immunology and Biochemistry of Hymenolepis diminuta 223 J. A NREASSEN, E.M. BENNET-JENKINS AND C. BRYANT Control Strategies for Human Intestinal Nematode Infections . . . . . . . . . . 277 M. A LBONICO, D.W.T. CROMPTON AND L. SAVIOLI DNA Vaccines: Technology and Applications as Anti-parasite and Anti-microbial Agents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 343 J.B. A LARCON, G.W. WAINE AND D.P. MCMANUS ADVANCES IN PARASITOLOGY VOL 62 ISSN: 0065-308X $35.00 DOI: 10.1016/S0065-308X(05)62014-3 Copyright r 2006 Elsevier Ltd. All rights of reproduction in any form reserved . March-May Figure 3 Top: changes in the annual numbers of cases of tick-borne encephalitis in Sweden, Hungary and Lithuania, 1960–2004. The step in- creases in Sweden from 1983 to 1986 and again in. sampling, 15–17, 30, 32 Brazil, yellow fever in, 184, 207–209 ADVANCES IN PARASITOLOGY VOL 62 ISSN: 006 5-3 08X $35.00 DOI: 10. 1016/S006 5-3 08X(05)6201 3-1 Copyright r 2006 Elsevier Ltd. All rights of. 307–308 Africa east African highlands, increased incidence of malaria in, 370–374 infectious disease contagion in, 377 malaria in, 355, 370–377 Rift Valley Fever (RVF) in, 16, 29 WNV in, 307, 322–323 yellow