The low power wide area network (LPWAN) is the bone of the Internet of Things. As the most successful wireless access technique for LPWAN, LoRa modulation has been widely applied to low mobility or static use cases. But can we use LoRa for high mobility applications, such as outdoor assets or animal tracking? We will answer this question by investigating the possibility of using LoRa in vehicular networks. In LoRa modulation, both propagation delay and Doppler shift can potentially lead to symbol detection errors. Based on our analysis, these two channel parameters are playing very similar roles in deteriorating system performance, and a linear combination of them can be used to quantify how good/bad the channel is. A fundamental tradeoff between signal resilience to delay and Doppler spreads is unveiled, which sheds light on waveform design for LoRa modulation in doubly dispersive channels. This observation is verified through simulations.