Risk maps for cities: Incorporating streets into geostatistical models

Area of Science:

• Spatial epidemiology
• Mathematical modeling

Background:

• Vector-borne diseases frequently emerge in urban settings.
• Gaussian field models are used for vector presence risk mapping.
• Existing models do not account for streets as barriers to insect movement.

Purpose of the Study:

• To develop a methodology for transforming spatial point data to incorporate permeable barriers, specifically streets.
• To estimate an additional parameter for Gaussian field models that accounts for street permeability.
• To create improved risk maps for vector-borne diseases in urban environments.

Main Methods:

• Developed a map transformation technique to widen streets, simulating permeable barriers.
• Utilized Gaussian field models to estimate a parameter quantifying street permeability.
• Applied the methodology to simulated datasets and real-world data of Triatoma infestans in Arequipa, Peru.

Main Results:

• The transformed landscape model, incorporating streets as permeable barriers, provided a better fit to observed Triatoma infestans infestation patterns.
• The best-fit model effectively doubled the perceived distance between houses separated by streets.
• Demonstrated the utility of the method for mapping vector distribution in urban areas.

Conclusions:

• The proposed methodology enhances spatial risk mapping for vector-borne diseases by accounting for urban infrastructure like streets.
• This approach offers a more realistic representation of vector dispersal in cities.
• Findings can inform more effective strategies for controlling urban insect vector populations and preventing disease transmission.