Fronts divide diazotroph communities in the Southern Indian Ocean

Abstract

Dinitrogen (N2) fixation represents a key source of reactive nitrogen in marine ecosystems. While the process has been rather well-explored in low latitudes of the Atlantic and Pacific Oceans, other higher latitude regions and particularly the Indian Ocean have been chronically overlooked. Here, we characterize N2 fixation and diazotroph community composition across nutrient and trace metals gradients spanning the multifrontal system separating the oligotrophic waters of the Indian Ocean subtropical gyre from the high nutrient low chlorophyll waters of the Southern Ocean. We found a sharp contrasting distribution of diazotroph groups across the frontal system. Notably, cyanobacterial diazotrophs dominated north of fronts, driving high N2 fixation rates (up to 13.96 nmol N l-1 d-1) with notable peaks near the South African coast. South of the fronts non-cyanobacterial diazotrophs prevailed without significant N2 fixation activity being detected. Our results provide new crucial insights into high latitude diazotrophy in the Indian Ocean, which should contribute to improved climate model parameterization and enhanced constraints on global net primary productivity projections.

Keywords: HNLC; N2 fixation; fronts; noncyanobacterial diazotrophs; subtropical gyre; trace metals.

Figure 1.

CTD profile and underway stations sampled during the cruise overlaid on (A) sea surface temperature (SST) and Chlorophyll-a composite of L3M 4 km product retrieved from the Copernicus marine service for the cruise period (13 January 2021 to 4 March 2021). STF = subtropical front, SAF = subantarctic front, and PF = polar front. (C) N₂ fixation (nmol N l⁻¹ d⁻¹) and (D) primary production (nmol C l⁻¹ d⁻¹) rates measured from surface samples (5 m) along the cruise transect. Nondetectable rates are depicted with crosses. AC = Agulhas current and ARC = Agulhas return current.

Figure 2.

Whisker plot of ¹⁵N cellular fractional abundance (atom%) for each group analyzed. Each dot represents a single analyzed cell. Gray dots denote cells with rates not significantly different from zero. Black lines denote mean ¹⁵N fractional abundance and standard deviations (horizontal plain and vertical dashed, respectively). ¹⁵N enrichments of heterotrophic bacteria were significantly different from Crocosphaera and Trichodesmium (Wilkoxon test, P < .001), but not Crocosphaera andTrichodesmium  ¹⁵N enrichments were not significantly different from each other (Wilkoxon test, P = .92).

Figure 3.

nifH gene abundance (log₁₀  nifH gene copies l⁻¹) of (A) Trichodesmium, (B) UCYN-A1, (C) UCYN-A2, (D) UCYN-B, (E) Gamma-A, and (F) Gamma-4 assessed by qPCR in surface samples (5 m) along the cruise transect.

Figure 4.

Heatmap representing most abundant ASVs across three temperature gradients for the (A) nifH gene, and (B) nifD gene. nifH groups have been reordered to facilitate comparison with nifD groups.

Figure 5.

Redundancy analyses showing the influence of environmental variables and N₂ fixation rates on diazotroph community composition based on (A and B) nifH genes and (C and D) nifD genes, grouped as north and south of the fronts or by temperature cluster, respectively.