The power of plant phenology monitoring networks - latitudinal patterns in phenology variance
Abstract
Large inter-annual variation in phenology monitoring data makes it a challenge to draw quantitative conclusions about phenological change as well as differences between species, phases and different places. We looked for geographical patterns in the standard deviation of phenology data [SD(onset)]. The data used was a sub-sample of a Swedish, nation-wide plant phenology database, covering a latitudinal range of 55-67°N, collected by an observation network between 1873-1917. Forty-four different species (perennial herbs, sub-shrubs, shrubs and trees) and four different phases (spring leaf-out, flowering, fruit ripening and autumn leaf coloration) were included, which thus covered the whole growing season (approx. Julian day 100-300). In addition to the commonly found decrease of the SD(onset) over the season, this data revealed a hump for late-summer flowering and fruit ripening species. SD(onset) was similar (and lowest) for spring leaf-out and autumn leaf coloration. The power analyses suggested that for about half of the species, including all but two spring leaf-out species, 9-11 years of present-day observations are needed to statistically infer a 7-day change in onset dates. In about a third of the species there was a correlation between SD(onset), and thus power, and latitude. The inter-specific comparison revealed a hump-shaped pattern, such that for early and late season phases SD(onset) decreased with latitude, while SD(onset) increased with latitude for mid-season phases.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2011
- Bibcode:
- 2011AGUFMGC43B0922B
- Keywords:
-
- 0410 BIOGEOSCIENCES / Biodiversity;
- 0439 BIOGEOSCIENCES / Ecosystems;
- structure and dynamics;
- 0476 BIOGEOSCIENCES / Plant ecology;
- 1630 GLOBAL CHANGE / Impacts of global change