No abstract

No abstract

To explore potential travel behavior shifts induced by personally owned, fully autonomous vehicles (AVs), we ran an experiment that provided personal chauffeurs to 43 households in the Sacramento region to simulate life with an AV. Like an advanced AV, the chauffeurs took over driving duties. Households were recruited from the 2018 Sacramento household travel survey sample. Sampling was stratified by weekly vehicle miles traveled (VMT), and households were selected to be diverse by demographics, modal preferences, mobility barriers, and residential location. Thirty-four households received 60 h of chauffeur service for 1 week, and nine households received 60 h per week for 2 weeks. Smartphone-based travel diaries were recorded for the chauffeur week(s), 1 week before, and 1 week after. During the chauffeur week, the overall systemwide VMT (summing across all sampled households) increased by 60%, over half of which came from “zero-occupancy vehicle” (ZOV) trips (when the chauffeur was the only occupant). The number of trips made in the system increased by 25%, with ZOV trips accounting for 85% of these additional trips. There was a shift away from transit, ridehailing, biking, and walking trips, which dropped by 70%, 55%, 38%, and 10%, respectively. Households with mobility barriers and those with less auto dependency had the greatest percent increase in VMT, whereas higher VMT households and families with children had the lowest. The results highlight how AVs can enhance mobility, but also caution against the potential detrimental effects on the transportation system and the need to regulate AVs and ZOVs.

Under scenarios of urbanization coupled with increasing frequency and intensity of natural hazards, urban disaster risk is set to rise. Simulating future urban expansion can provide relevant information for the development of future exposure scenarios and the identification of targeted risk reduction and adaptation strategies. Here, we present an urban growth simulation for the coastal city of Monastir, Tunisia. The approach integrates local knowledge and a data-driven urban growth model to simulate urban sprawl up to 2030. A business-as-usual projection is used to predict the future growth of the city based on the historical trend. Thirteen Landsat images for the period 1975 to 2017 were used to delineate past changes in urban land cover following the European Urban Atlas standard, which served as the main input for the urban growth model. The simulation revealed that the city's residential area is likely to grow by 127 ha to an overall size of 1,690 ha by 2030, corresponding to an increase of 8.1% compared to the urban footprint of 2017. The outcomes of the analysis presented here served as an input for the spatial simulation of future exposure to flash floods in the case study area.

No abstract