Development and analysis of spring plant phenology products: 36 years of 1-km grids over the conterminous US

摘要

Time series of phenological products provide information on the timings of recurrent biological events and on their temporal trends. This information is key to studying the impacts of climate change on our planet as well as for managing natural resources and agricultural production. Here we develop and analyze new long term phenological products: 1 km grids of the Extended Spring Indices (SI-x) over the conterminous United States from 1980 to 2015. These new products (based on Daymet daily temperature grids and created by using cloud computing) allow the analysis of two primary variables (first leaf and first bloom) and two derivative products (Damage Index and Last Freeze Day) at a much finer spatial resolution than previous gridded or interpolated products. Furthermore, our products provide enough temporal depth to reliably analyze trends and changes in the timing of spring arrival at continental scales. Validation results confirm that our products largely agree with lilac and honeysuckle leaf and flowering onset observations. The spatial analysis shows a significantly delayed spring onset in the northern US whereas in the western and the Great Lakes region, spring onset advances. The mean temporal variabilities of the indices were analyzed for the nine major climatic regions of the US and results showed a clear division into three main groups: early, average and late spring onset. Finally, the region belonging to each group was mapped. These examples show the potential of our four phenological products to improve understanding of the responses of ecosystems to a changing climate.