Using digital camera images to analyse snowmelt and phenology of a subalpine grassland

Plant phenology is a commonly used and suitable indicator of the impact of climate change on vegeta-tion. In mountainous areas, phenology is governed by environmental drivers such as air temperature,photoperiod and the presence of snow. In this study, digital images collected over 3 years (2009, 2010and 2011) in a subalpine grassland site were used to investigate the relationship between the timing ofsnowmelt and the beginning of the growing season in both the spatial and the temporal dimension.The image analysis was conducted for a wide area corresponding to approximately 150 m2to char-acterize the spatial heterogeneity of grassland phenology. The investigated area was divided into 85510 × 10 pixel cells, and for each cell annual time series of green chromatic coordinates (gcc) were com-puted from hourly images. To analyse the spatial pattern of phenology, the beginning of the season foreach cell was extracted from the gcc time series. Based on the same grid dimension, three maps of yearlysnowmelt date corresponding to the day of the year in which the snow in each cell disappeared from theground were obtained.Although complete snowmelt in the area occurred rapidly, within a maximum of six days, several dis-tinct spatial patterns were identified with snowmelt occurring earlier in convex compared to concaveareas. Differences in snowmelt dates were quite unexpectedly negatively related to the beginning of thegrowing season. The negative correlation was explained considering that areas characterized by differentmicrotopography have also a different species composition: the growing season began earlier in concaveareas preferred by opportunistic species with a fast development after snowmelt while phenologicaldevelopment of grass typical of convex areas can take longer. This behaviour was especially evident in2011 characterized by an extremely anticipated snowmelt. On the contrary, the analysis of the relation-ship between the timing of snowmelt and the beginning of the season between the three years analysedin this study, highlighted an advancement of the beginning of the growing season in 2011. However, thisis valid only in areas characterized by the abundance of opportunistic species such as forbs for which thesnow cover plays a major role in determining the beginning of phenological development. The resultspresented in this study support the possibility of using repeat digital photography to analyse the role ofplant species composition on phenology in complex ecosystems such as subalpine and alpine grasslands