2.2 Observed event Ewc estimates from forest plot studies in temperate locations
Studies relevant to UK conditions with forest plot CWB Ewcobservations were identified. Relevant studies were defined climatologically using the revised Köppen climate classification (Chen & Chen, 2013) with the recognition that the classification provides general climatic classes within which there is a large degree of variability. Although this variability exists, Köppen climate classifications were used with the rationale that hydrometerological conditions will have similarities to those of UK mountainous regions during large and extreme rainfall events. The specific classifications deemed acceptable for the purposes of this study were: Cfa (mild temperate; fully humid; hot summer), Cfb (mild temperate; fully humid; warm summer), Cfc (mild temperate; fully humid; cool summer), Csb (mild temperate; dry summer; warm summer), Dfb (snow; fully humid; warm summer), Dfc (snow; fully humid; cool summer), Dfd (snow; fully humid; cold summer): the classifications Cfb and Cfc encompass all mountainous regions of the UK. For the climate classes where snow can form a significant part of winter precipitation (i.e. Dfb, Dfc and Dfd), care was taken not to include data affected by snow falls: in most cases this was already carried out in the original study.
From the studies identified as relevant (Table Supp. 1), only CWBEwc observations reported on a rainfall event basis or as a daily total were collated for analyses. For consistency, events reported to be over 24 hours (but no more than 48 hours) duration were standardised by calculating a normalised 24-hour rainfall (i.e.\(Pg=\ 24\left(\frac{\text{Pe}}{\text{De}}\right)\)) wherePe is the total event precipitation and De is event duration in hours): note that, in the text below, 𝑃𝑔 relates to a daily or normalised daily total rainfall unless otherwise stated. It is accepted that using daily observations is somewhat artificial in as much as the duration of rainfall events may be truncated where they span multiple days; it is also recognised that both daily and event data may include periods without rain. Event-based data where event duration was not reported were also collated and are presented separately.
Data were gathered using values provided in tables or by digitising data presented as figures in published material and by abstracting data from field log sheets; in the case of Aussenac (1968) and Reynolds and Henderson (1967) Pg-Ewc relationships were digitised. Where observations were obtained from digitised figures, and where there were many data points, obscured or overlapping data will have resulted in some values not being included; these data were, however, invariably for relatively low 𝑃𝑔 magnitudes. In some cases, the dates associated with individual rainfall events were not provided in the published material, which does not allow separation by season (or by leafed or leafless period for deciduous species). Furthermore, as event data were limited and data for deciduous forest plots were very few, all events irrespective whether they were evergreen or deciduous were combined for the comparison with theoretical estimates, but are identified separately in figures.