Analyzing the Patterns of Space-Time Distances for Tracking the Diffusion of an Epidemic - Space-Time Integration in Geography and GIScience - page 271

Fig. 15.5 Locations of Kaohsiung and Fengshan, Taiwan

The data on dengue cases was collected through the national surveillance
system of Taiwan Centers for Disease Control (Taiwan CDC), which reported 785
confirmed dengue cases (DEN) in the two-city area from July 2009 to February
2010. The epidemic curve in Fig. 15.6 shows two significant epidemic waves.
The period of the initial wave was from July to late August. The second wave
(the major epidemic) lasted from early September to February of the next year.
To understand the progression of the epidemic, we divided this epidemic into five
periods: the initial wave (period 1, July-late August), pre-epidemic of the second
wave (period 2, early September-October), major epidemic of the second wave
(period 3, November), post-epidemic of the second wave (period 4, December), and
January- February of the next year (period 5).
Figure 15.7 shows the cumulative pairs of space-time distances for the dengue
epidemic. Deeper red coloring indicates the presence of more cumulative pairs in a
given space-time distance interval, which also indicates spatial-temporal clustering
or interaction. The space-time clusters were also identified by outlier detection and
are represented as red dots. Four space-time clustering patterns were identified.