Assessment of seasonal and decadal signals in the in-situ Atlantic plankton data

No abstracts are to be cited without prior reference to the author.

Combining in-situ plankton time series data sets from different research cruises over an oceanAtlantic scale is particularly challenging. Specifically creating a knowledge base that enables statistical analysis of seasonal and the longer term structure and function of the planktonic presents two challenges: 1) subsampling: the underlying trends are not completely represented as not all measures are regularly sampled over time (months & years), and 2) measurement protocols: scientific protocols for individual research expeditions need to be harmonised so that representative long term time series can be constructed. We present the current learning from combining in-situ data across Arctic (1913+), Atlantic (1985+) and Southern Ocean (1992+) waters. To help address these challenges, this in-situ data and the related knowledge is being combined with biogeochemical numerical models and remote sensing data in the GreenSeas Analysis Framework. This novel framework will be briefly discussed to explain how the GSAF and its linkage with the insitu sampling design provide a systematic capability to construct the longer term time series. Ultimately it enables analyses which help improve the numerical models which are used to give future assessments of climate. We report here the current learning obtained from combining these in-situ data with a focus on phytoplankton abundance, primary production and nutrient metrics. We discuss the seasonal and decadal trends in this data, and present (with relevant statistically simulated data) methods to distinguish seasonal and decadal variation from longer term trends that are typically present in these types of time series.