Figure 2. Variations of the gravity model were fit to data to capture different types of trips.

Here, Kenya is used to demonstrate the trip types that could be defined by the region (represented by color) and/or urbanicity (solid = predominantly rural, dotted = predominantly urban) of the trip’s origin and destination to outline the various models fit. See Figure 2—figure supplements 1–2 for the fitted parameters.

Figure 2—figure supplement 1. Gravity model parameters (power distance kernel).

Mobile phone data was used to fit parameters for origin population size (α), destination population size (β), and trip distance (γ) for each country. Different sets of parameters were fit, based on the gravity model type (color) and trip type (I: intra-regional, O: inter-regional, U: urban, R: rural). While the general trend of distance parameter as a function of trip type was similar across countries, the role the population size parameters played was country-specific. Note that there were no IUU routes in Namibia nor UU, IRU, IUR, IUU, or OUU routes in Burkina Faso, resulting in large standard deviations for the fitted parameters. Parameter values presented in Figure 2—source data 1.

Figure 2—figure supplement 2. Gravity model parameters (exponential distance kernel).

Mobile phone data was used to fit parameters for origin population size (α), destination population size (β), and trip distance (D) for each country. Different sets of parameters were fit, based on the gravity model type (color) and trip type (I: intra-regional, O: inter-regional, U: urban, R: rural). While the general trend of distance parameter as a function of trip type was similar across countries, the role the population size parameters played was country-specific. Note that there were no IUU routes in Namibia nor UU, IRU, IUR, IUU, or OUU routes in Burkina Faso, resulting in large standard deviations for the fitted parameters. Parameter values presented in Figure 2—source data 1.