Climate change and sea level rise have increased the frequency and severity of flooding events in coastal communities. This study quantifies transportation impacts of recurring flooding using crowdsourced traffic and flood incident data. Agency-provided continuous count station traffic volume data at 12 locations is supplemented by crowd-sourced traffic data from location-based apps in Norfolk, Virginia, to assess the impacts of recurrent flooding on traffic flow. A random forest data predictive model utilizing roadway features, traffic flow characteristics, and hydrological data as inputs scales the spatial extent of traffic volume data from 12 to 7736 roadway segments. Modeling results suggest that between January 2017 and August 2018, City of Norfolk reported flood events reduced 24 h citywide vehicle-hours of travel (VHT) by 3%, on average. To examine the temporal and spatial variation of impacts, crowdsourced flood incident reports collected by navigation app Waze between August 2017 and August 2018 were also analyzed. Modeling results at the local scale show that on weekday afternoon and evening periods, flood-impacted areas experience a statistically significant 7% reduction in VHT and 12% reduction in vehicle-miles traveled, on average. These impacts vary across roadway types, with substantial decline in traffic volumes on freeways, while principal arterials experience increased traffic volumes during flood periods. Results suggest that analyzing recurring flooding at the local scale is more prudent as the impact is temporally and spatially heterogeneous. Furthermore, countermeasures to mitigate impacts require a dynamic strategy that can adapt to conditions across various time periods and at specific locations.

气候变化和海平面上升增加了沿海社区洪水泛滥的频率和严重性。这项研究使用众包交通和洪水事件数据量化了反复洪水的运输影响。由机构提供的12个位置的连续计数站交通量数据得到了弗吉尼亚州诺福克基于位置的应用程序的众包交通数据的补充，以评估经常性洪水对交通流量的影响。利用道路特征，交通流量特征和水文数据作为输入的随机森林数据预测模型可将交通量数据的空间范围从12到7736个道路段定标。建模结果表明，2017年1月至2018年8月之间，诺福克市报告的洪水事件平均使全市24小时车辆旅行小时数（VHT）平均降低了3％。为了检查影响的时间和空间变化，还分析了导航应用Waze在2017年8月至2018年8月之间收集的众包洪水事件报告。在当地范围内的建模结果表明，在工作日下午和晚上，受洪灾影响的地区的VHT平均降低了7％，而行车里程平均降低了12％。这些影响因道路类型而异，高速公路上的交通量大幅下降，而主要干道在洪水期间交通量有所增加。结果表明，由于影响在时间和空间上是非均质的，因此在局部范围内对重复性洪水进行分析更为谨慎。此外，