The “ M elosira years” of Lake B aikal: Winter environmental conditions at ice onset predict under‐ice algal blooms in spring

• Published in: Limnology and oceanography Vol. 60; no. 6; pp. 1950 - 1964
• Main Authors: Katz, Stephen L., Izmest'eva, Lyubov R., Hampton, Stephanie E., Ozersky, Ted, Shchapov, Kirill, Moore, Marianne V., Shimaraeva, Svetlana V., Silow, Eugene A.
• Format: Journal Article
• Language: English
• Published: 01.11.2015
• ISSN: 0024-3590; 1939-5590
• DOI: 10.1002/lno.10143

Abstract Winter primary production in seasonally ice‐covered lakes historically has not been well studied, but it is increasingly recognized as an important component of lake metabolism. Lake Baikal in Siberia is not only the World's oldest, deepest, and most biologically diverse lake, but also where large under‐ice blooms of the diatom Aulacoseira baicalensis (formerly Melosira ) occur in some years. The phenomenon of “Melosira years” is noteworthy both for the intensity of the diatom blooms, in which total under‐ice production can be a majority of total annual production, and for the enigmatic regularity of their occurrence every 3–4 yr. The degree to which these episodic blooms might be controlled by external forcing and endogenous lake processes has been debated for decades. We used a 50‐yr time series of phytoplankton observations to statistically model the occurrence of Aulacoseira blooms as a function of meteorological and climatological predictor variables. The results support the hypothesis that a confluence of meteorological conditions in the preceding fall season, which favor clear ice formation with minimal snow cover, also favor Aulacoseira blooms in the following spring. Further, we observe that this confluence of factors is related to relatively strong states of the Siberian High which, while not strictly periodic, do explain a significant fraction of the interannual bloom pattern. Finally, our analyses show that the timing of the peak abundance of A. baicalensis shifted 1.6 months later across the 50‐yr time series, corresponding with the delay in ice‐on timing that has been associated with climate change.