Optimizing COST Action 733 Weather Type Classifications over Eddy Covariance Fluxes at Collelongo Beech Forest eLTER site (Italy)

Scientific evidence shows that climate and meteorological conditions strongly modulate forest CO₂ fluxes measured using Eddy-Covariance method. However few studies have investigated the impact of synoptic-scale circulation on gas exchange variability.

Seasonal estimates of Gross Primary Production (GPP), Net Ecosystem Exchange (NEE) and Ecosystem Respiration (RECO) were derived using linear regression models applied to Eddy Covariance data collected at the Collelongo-Selva Piana LTER-Italy site (3000 ha beech forest) from 1996 to 2014. Circulation Weather Types (CWTs) derived from ERA5 MSLP and HGT500 (used individually and combined) are generated via principal methods available in the used software (Simulated Annealing, Leader, Principal Component Analysis, Threshold methods).

We use COST Action 733 software to construct and compare multiple CWT classifications over a fixed Italian domain, selecting the optimal classification for explaining CO₂ flux variability.

Classifications are ranked by minimizing intraclass and maximizing interclass variance of GPP, NEE and RECO. This approach identifies the CWT classification that best discriminates flux regimes driven by synoptic patterns (uptake under westerlies vs. emissions under blocking).

The optimized Italian CWT enables diagnostic attribution of observed CO₂ anomalies to specific circulation drivers, helping to clarify Collelongo beech forest ecosystem processes.

In addition, the CO₂‑optimized CWT classification could be applied in CORDEX climate scenarios to quantify the future impacts of changing circulation frequencies on forest ecosystem activities.

Hence, understanding how CWTs influence natural ecosystem fluxes — and how these circulation patterns may evolve under future scenarios — could provide valuable guidance for forest management. This includes strategic species selection for adaptive silvicultural practices, supporting decision makers in climate-resilient forestry planning.