Continental-Scale Monitoring and Mapping of False Spring: A Cloud Computing Solution

Global warming has shifted the onset of spring plant phenology towards earlier dates. This shift can lead to “false springs” (plants get frost damage) because the date of last frost has not advanced at the same pace than the advancement of spring onset. Here, we use a cloud computing approach for processing big and high spatial resolution grids of temperature data to map the occurrence of false springs and to study their temporal trends. We demonstrate our approach using Daymet, which provides daily weather data at 1km for continental US, and the extended spring indices models. Results show that the risk of having false springs is high in the Midwest and in Northeast of the US. This extensive regional variation highlights the necessity of continuous monitoring and mapping of false spring.