Analyzing the Impacts of Seasonal Climate Change on Forest Phenology in Maine Across Two Decades

Abstract:

Maine is an important yet unique location to study the impacts of climate change on forest phenology because it is situated in a transition zone between eastern temperate forest to the south and boreal forest to the north resulting in more pronounced changes than are seen in most other U.S. states. In this study, we used remote-sensing techniques to analyze the time-series of day-of-year occurrence for 8 phenology metrics (e.g. greenup, peak, dormancy) from 2001 to 2017 among 7 biophysical regions within the state of Maine, U.S. Imagery was extracted from the MODIS MCD12Q2 Land Dynamics Collection at a 500-m resolution. Results show that most change occured from greenup through peak and length of season (LOS), with the northern and mountainous regions of Maine exhibiting greater changes relative to southern and coastal regions. In addition, the average LOS in Maine has extended by 9.7± .29 days (p=.052), with a range between 11.8± .32 days (p< .05) for the Aroostook Hills & Lowlands biophysical region and 2.4± .31 days (p=.64) for the Southern Maine region over the 16 year time period. Variable Importance Using Random Forest models were used to identify the most important seasonal climate variables impacting each phenology metric at both a statewide- and biophysical regional-scales. Preliminary results show that the accumulated growing degree days has a major influence for the observed change within Maine at spring greenup with a threshold of 0°C. The findings from our research provide a better understanding of the impacts of climate change on forest ecosystem dynamics and can be used to predict future phenological trends at different spatial scales globally.