• No results found



associated temperature change of 41.0 to 69.1 °C for this feature during overflow (Figure C.4;

Table C.6). These Thermoproteota signatures were subsequently evidenced in downstream sites of the overflow, decreasing in relative abundance from 12.5 % of the total microbial community in Waimangu Feature 4 to 0.1 % in Waimangu Feature 7. Other variable phyla across Waimangu Stream included increases in Pseudomonadota (not Acidithiobacillus sp.) in all sites except Feature 4, and decreases in Cyanobacteria in all sites. Features 6 and 7, which were near the stream outflow into Lake Rotomahana, also had increases in Bacteroidota, which corresponded with decreases in obligately thermophilic Aquificota. The effect of Inferno Crater Lake overflow was apparent in the physicochemistry of features immediately downstream of the overflow entry point (Figure C.4; Table C.6), where pH decreased from 6.7 to 3.0 in Feature 4, and 7.4 to 3.0 in Feature 5. Temperature fluctuation in downstream sites was less apparent, with the greatest temperature increase due to overflow (+6.7 °C) occurring in Waimangu Feature 4. Interestingly, alpha diversity analyses did not reflect the physicochemical results, with the greatest changes observed in OTU richness of Waimangu Feature 4 and Feature 6 (Figure C.4; Table C.6).

Despite a complete shift in microbial community structure in Inferno Crater Lake between the two timepoints, both alpha diversity measures for this ecosystem had negligible variation, with an associated temperature difference of 28.1 °C (Figure C.4; Table C.6). The greatest dissimilarity in beta diversity between the temporal stream samples was found in Features 6 and 7 (Figure C.4). Even with the pH and temperature changes in Features 4 and 5 post-disturbance, beta diversity did not show variance between the microbial communities of these features over time.

3.5.3 Diversity metrics showed initial response to long-term anthropogenic disturbance In contrast to the short-term, natural disturbance observed at Waimangu, variations in alpha and beta diversity of microbial communities did correspond with a long-term, anthropogenic disturbance at the Waikite wetland, in particular for sites located immediately before and after the installation of a weir (Features 6 and 8 respectively; Figure 3.2; Table C.7). In all wetland sites, OTU richness and evenness then returned to pre-disturbance levels for the final sampling timepoint (23 months post-weir installation). Beta diversity showed an initial response to the disturbance for all sites in or near Otamakokore Stream, which was then amplified for the final timepoint of Features 3, 4 and 8. Features with an additional geothermal input (Feature 5 and


‘Pig Scorcher’) demonstrated variability in beta diversity irrespective of the disturbance, which was also apparent in the relative abundances of community structure in these sites (Figure 3.4;

Table C.7). For example, the ‘Pig Scorcher’ geothermal spring had a notable decrease in Deinococcota before the disturbance was introduced to the wetland, which coincided with an increase in Aquificota. Additionally, the microbial community of the ‘Big Spring’ exhibited fluctuations in relative phyla abundances before and after the weir installation. However, as only two timepoints were sampled for this site, the question of whether these changes were associated with the disturbance remain unresolved. While the resident taxa of microbial communities upstream to the wetland (Features 3 and 4) remained constant over time, increases in the relative abundances of Pseudomonadota, coupled with decreases in Cyanobacteria and Bacteroidota, did correspond with the weir installation (Figure 3.4). Similarly, the relative abundances of Pseudomonadota increased, while Bacteroidota and Verrucomicrobiota decreased in Waikite Feature 6, a site within the wetland, although compositional changes not related to the disturbance were also noted in the microbial community of this feature. The outlet to the wetland (Feature 8) showed increases in Parcubacteria and Chlamydiota at the first timepoint post- disturbance, but these reduced to similar levels as the first timepoint by the end of the experiment. Overall, Pseudomonadota increased and Bacteroidota decreased at the wetland outlet across the three years analysed. Regarding physicochemistry, all features in and around the wetland decreased in both pH and temperature from December 2013 to October 2016 (Figure 3.2; Table C.7). The majority of these changes were minor (pH: SD≤0.2; temperature: SD≤4.5

°C), except for Waikite Features 6 and 8 which had standard deviations of pH 0.4 and 13.5 °C, and pH 0.6 and 10.3 °C respectively. Despite these variations, no significant correlations (p<0.05, Spearman’s coefficients) were observed between diversity and temperature for all features, with only beta diversity of Waikite Features 4 and 8 producing meaningful correlations with pH (Figure C.5).


Figure 3.4 - Temporal microbial community composition and relative abundance of Waikite geothermal features. Amplicon sequencing of the 16S rRNA gene was used to measure read abundance of taxa in spring communities, with only phyla >1 % average relative abundance across all samples in this category shown. The vertical dashed line indicates the construction of a weir near the outlet of the wetland during restoration.


3.5.4 Temperature is more variable than pH over time in geothermal features

Similar to control and disturbed samples, temporal variations were observed in the relative abundances of geothermal microbial communities across a range of pH values (pH 3 to 9; Figure 3.5; Table C.8). Six of the original 69 samples did not yield sufficient sequence reads (<5,000) after rarefaction, resulting in 63 samples for final community analysis. The greatest changes to relative abundances occurred in two features from the pH 3 group (Whakarewarewa Features 51 and 53), and two features from the pH 9 group (Whakarewarewa Feature 1 and Ohaaki Feature 2). Aquificota taxa were involved in all these changes, with relative abundance of the phylum increasing in Whakarewarewa Feature 51, but decreasing in Whakarewarewa Feature 53 over the timeline of the experiment. These variations were coupled with decreasing Thermoproteota (family Sulfolobaceae; Figure C.6) and increasing Pseudomonadota respectively, while the balance of Aquificota in the two pH 9 features were inversely proportional to Deinococcota taxa.

Aquificota was also the most abundant phylum in five of six features from the pH 5 and 7 groups (as genus Venenivibrio; Figure C.6), with notable variation to this phylum only occurring in Kuirau Park Feature 60. Whangapoa Feature 1 (pH 7 group) was the only feature across all 12 sites in this category to present change independent of Aquificota; here, relative abundances of Armatimonadota, Bacteroidota, and Deinococcota fluctuated over the 10-months analysed.

Alpha and beta diversity metrics for all 12 features in this category produced contradictory results (Figure 3.6; Table C.8), with the greatest variation in alpha diversity presented in the pH 7 group features. Conversely, beta diversity suggested geothermal features from pH 3, 5 and 9 groups were more likely to have varied community structure over time, and was more representative of the variation in relative abundances than alpha diversity. pH remained relatively stable over time for the majority of features, with only Whakarewarewa Feature 51 (pH 3 group) having a standard deviation >0.3 pH units (Figure 3.6; Table C.8). Analysis of similarities confirmed that variation in beta diversity was greater between pH groups than within pH groups (p=0.001, ANOSIM; Figure C.7), indicating the significant influence of pH on driving diversity in these ecosystems on a spatial scale. However, temperature exhibited a greater temporal variability than pH, with only four features having a standard deviation of ≤2 °C (Figure 3.6; Table C.8). Whakarewarewa Feature 1 (pH 9 group) had the greatest fluctuation in temperature overall, changing from 63.4 to 80.6 °C, over the sampling period. The effect of


temperature on these ecosystems was highlighted by six features producing significant positive correlations between temperature and beta diversity (p<0.05, Spearman’s coefficient; Figure C.8). The physicochemical relationship with alpha diversity was not as conclusive, with only four features demonstrating significant correlation between temperature and either OTU richness or evenness (p≤0.03, Spearman’s coefficient; Figure C.8). Corresponding with the pH stability we observed in these geothermal features, Whakarewarewa Feature 51 (group pH 3) was the only feature to produce a positive correlation between pH with both alpha and beta diversity metrics (OTU richness, ρ=0.70, p=0.03; Bray-Curtis dissimilarities, ρ=0.84, p=0.01; Spearman’s correlation coefficient).

Figure 3.5 - Temporal microbial community composition and relative abundance of geothermal features across the pH scale. Amplicon sequencing of the 16S rRNA gene was used to measure read abundance of taxa in spring communities, with only phyla >1 % average relative abundance across all samples in this category shown. Geothermal features are grouped according to their respective pH group (pH 3, 5, 7, and 9).



Figure 3.6 - Temporal diversity and physicochemistry of geothermal features across the pH scale.

(a) The microbial communities of 12 geothermal springs were measured six times over one year by amplicon sequencing of the 16S rRNA gene, with variation in alpha diversity between temporal samples indicated by OTU richness (i.e., the number of OTUs per community). (b) Alpha diversity was also measured by the Shannon diversity index to indicate evenness of the communities over time.

(c) Spring pH of the features at the time of sampling. (d) Spring temperature of the geothermal features. (e) Beta diversity of the 12 features is shown by a non-metric multidimensional scaling (NMDS; n=62, stress=0.15) of Bray-Curtis dissimilarities between all temporal samples, with data ellipses generated from multivariate t-distribution with a 95% confidence level. All samples are coloured according to feature name, with alpha diversity and physicochemistry further grouped by pH group.