Antarctic lake phytoplankton and bacteria from near-surface waters exhibit high sensitivity to climate-driven disturbance

Abstract

The McMurdo Dry Valleys (MDVs), Antarctica, represent a cold, desert ecosystem poised on the threshold of melting and freezing water. The MDVs have experienced dramatic signs of climatic change, most notably a warm austral summer in 2001-2002 that caused widespread flooding, partial ice cover loss and lake level rise. To understand the impact of these climatic disturbances on lake microbial communities, we simulated lake level rise and ice-cover loss by transplanting dialysis-bagged communities from selected depths to other locations in the water column or to an open water perimeter moat. Bacteria and eukaryote communities residing in the surface waters (5 m) exhibited shifts in community composition when exposed to either disturbance, while microbial communities from below the surface were largely unaffected by the transplant. We also observed an accumulation of labile dissolved organic carbon in the transplanted surface communities. In addition, there were taxa-specific sensitivities: cryptophytes and Actinobacteria were highly sensitive particularly to the moat transplant, while chlorophytes and several bacterial taxa increased in relative abundance or were unaffected. Our results reveal that future climate-driven disturbances will likely undermine the stability and productivity of MDV lake phytoplankton and bacterial communities in the surface waters of this extreme environment.

FIGURE 1

(A) Schematic of transplant experiment. Natural communities residing at various depths in the water column (5, 17 and 23 m) were transferred to dialysis bags, attached to PVC frames, and transplanted to either the open water moat (tMICE) or 3 m deeper in the water column (tLICE). Frames were retrieved after 14 days incubation. (B) PVC frame design. (C) A frame containing transplanted microbial communities

FIGURE 2

(A) Long‐term variations in ice thickness and lake level in Lake Bonney (shown as meters above sea level, masl), east lobe. (B) Long‐term trends in depth integrated under ice PAR (5–22 m) during December 1993–2019. Dotted line shows the year of the flood year. Abbreviation: PAR, photosynthetically available radiation

FIGURE 3

Long‐term trends in Chl a levels in Lake Bonney, east lobe. (A) Depth profiles. (B) Integrated levels in the water column during the months of November and December. Dotted line shows the year of the flood year.

FIGURE 4

Community composition and distribution of four major spectral algal classes in Lake Bonney, east lobe during the austral summer. Measurements were performed using a bbe FluoroProbe fluorometer during early summer (A,D), mid‐summer (B,E) and late summer (C,F). Samples for the transplant experiment were collected during early January 2018.

FIGURE 5

Phytoplankton community composition (A) and maximum photosynthetic efficiency (B) in response to mimicked lake level rise and moat expansion. Communities were transplanted in dialysis bags from either 5 m under‐ice layer to the open water moat (tMICE, left) or 3 m deeper in the water column (tLICE, right). Algal groups shown in different colours. Chlorophyll a was measured using the Phyto‐PAMII from TFF concentrated samples (n = 3)

FIGURE 6

Changes in relative abundance in eukaryotic (A) and bacterial (B) communities in response to mimicked moat expansion (tMICE, left) lake level rise (tLICE, right). Bacteria and eukaryote diversity was determined by 16S rDNA and 18S rDNA sequencing, respectively (n = 2–3)

FIGURE 7

PCoA cluster analysis of 18S rDNA (A) and 16S rDNA(B) from Lake Bonney (east lobe) in response to mimicked moat expansion or lake level rise. Circles, natural communities; squares, transplanted communities (n = 1–3). The scatter plots are of principal coordinate 1 (PC1) versus principal coordinate 2 (PCs) from weighted UniFrac results (bidirectional arrows show homogenization of communities; unidirectional arrows represent sensitivity of shifted communities). Abbreviation: PCoA, principal coordinate analysis

FIGURE 8

Changes in relative abundance in eukarya (A) and bacteria (B) communities in response to light treatment or nutrient amendment. Bacteria and eukaryote diversity was determined by 16S rRNA and 18S rRNA sequencing, respectively (n = 3). D0, Day 0; LL, low light (12 μmol m⁻² s⁻¹); ML, medium light (36 μmol m⁻² s⁻¹); HL, high light (190 μmol m⁻² s⁻¹) + N, +20 μM NH₄Cl; +P + 2 μM KH₂PO₄; +NP, +20 μM NH₄Cl/+2 μM K H₂PO₄