2.2.1. Precision mapping of active DF zones: a KDE-Based cross-sectional survey approach

Fig. 2a–c shows kernel density for each year (2017–2019), and Fig. 2d shows the spatiotemporal magnitude of change from 2018 to 2019 and the common risk areas. Rawalpindi city (33° 3433.50 N 73° 333.74″ E) is the most dengue-affected area and the best selection for our cross-sectional survey. The survey follows Rawalpindi's hotspots in a major urban zone (Fig. 1; 2,097,824 households). The sample area is the most populous urban zone with densely built-up land (Fig. 1). In the main urban zone, major DF concentrations clustered differently each year. In 2017, DF stayed in Chaklala Scheme 3, Gulzar-e-Quaid, Bilal Colony, and Dhoke Chaudrian. However, Allama Iqbal Colony and its neighbors experienced a pattern shift in 2018. In 2019, the main clusters were surrounding Gulzar-e-Quaid. Since the incident density in 2019 is much higher than in 2018, the difference map illustrated the same areas (Fig. 2d). When confirmed DF cases exceeded 49, a neighborhood was considered at risk. Given such major hotspot areas and their neighborhoods, survey locations were randomly selected based on a proximity scale, giving more importance to the main hotspots and decreasing their influence as distance increases. By aligning samples according to incidence, epidemic-hit communities can be targeted for an effective cross-sectional survey (Fig. 3).

KDE illustrating DF (confirmed cases) magnitude per area (2017–2019). (a) Focused 2017, (b) Focused 2018, (c) Focused 2019 and (d) Focused changes

Map showing KDE-based spatial isolation (criteria) of potential sample sites for KAPs survey in Rawalpindi metropolitan (Imagery Source: Esri, Maxar, GeoEye, Earthstar Geographics, CNES/Airbus DS, USDA, USGS, AeroGRID, IGN, & the GIS User community).