Palaeo-flux records from electron microscope studies of Holocene laminated sediments, Saanich Inlet, British Columbia

Abstract

Holocene varved sediments recovered from Saanich Inlet, ODP Leg 169S, have been analysed using high-resolution scanning electron microscope techniques. The individual centimetre-scale varves form couplets, comprising diatom ooze/diatomaceous mud deposited during spring to autumn and silty clay deposited during winter. Each individual varve may contain up to 19 laminae, recording sub-seasonal to seasonal-scale processes. These intra-annual laminae contain a range of components, including pelagic faecal pellets and aggregates of diatomaceous material. The combination of these individual components in varying proportions with a heterogeneous sediment, leads to a range of laminae fabrics. These include intact monospecific Skeletonema costatum and Thalassiosira pacifica diatom oozes and highly fragmented, aggregate- and faecal pellet-dominated diatom oozes which are typically Chaetoceros spp.-dominated. Additional laminae fabrics include diatomaceous mud, stringer-type diatomaceous mud which contains discrete diatom aggregates, homogeneous and pelleted silty clay. The varved diatom succession typically commences in early spring with Thalassiosira spp./Chaetoceros spp., followed by Chaetoceros spp. in late spring. During late spring/early summer blooms of S. costatum may be recorded, and summer to autumn production is characterised by one to several Chaetoceros spp.-dominated blooms. Inter-annual variability in diatom flux typically comprises the presence/absence of Thalassiosira spp. and S. costatum. In addition, resting spores of Chaetoceros spp., Thalassiosira nordenskioeldii and Leptocylindricus danicus are intermittently recorded. There is a major change in varve thickness from thinner varves prior to approximately 2100 yr BP to thicker varves after this date, which is attributed to increased terrigenous input throughout the varve. The identification of laminae fabrics, diatom successions and a unique massive interval has enabled an inter-site correlation between sites 1033B and 1034B.