Unnamed Article

Abstract
Double click to add an Abstract
Paper outline
Exploring potential of NGS of foraminfera for use in biotic indices for the purpose of characterising benthic enrichment
Escalating labour costs restricting use of macrofauna for discerning effects – less replication, less stations, less frequent. This paper uses widely utilised and accepted indicators of benthic enrichment to identify ecological characteristics, allocate Eco-Groups (Followed methods of Keeley et al. 2012) in a manner analogous to established macrofauna methods (Borja et al). Then this information is used to apply a variety of established indicators to explore the potential for discerning the enrichment gradient.
Comment on transformations and number of reads - abundance.
Comment on best indices.
Comment on how well best indicator did the job.
NGS formanifera data can be used to calculate new biotic index which reliable discerns enrichment stage. Cheaper faster, better. Works well now, lots of room for further refinement.
  1. Summary of foram and standard data (refer to Pochon et al.)
  2. Describe ES process – refer Keeley et al. 2012a,b.
  3. Describe splines and characteristics
  4. Allocation of preliminary EG’s – Using MS Y1 data
  5. Testing biotic indices –
    1. Compare transformations – relationship to ES
    2. Chosen index with standard indicators and biotic indices including ES
    Hi,
    Here’s some figures that will hopefully get you a little excited again. I tried to put this down and do some other work, but I couldn’t, so I have pushed on and got to a logical rest point.
    In short what I did was:
    · Filtered the previous EG allocations from Year1 data to a select few and applied them to the Y2 Sounds and BGB data. Sorted and standardized the new data and then calculated the fAMBI’s for all the datasets and constructed some code to plot them up. I lost a lot of information due to the inability to cross-over OTU’s and the resulting regressions did not look great – bit despondent at this point. I won’t bother sending those plots.
    · I think this was also to do with the fact that I had allocated EcoGroups (EG’s) based on one dataset, and as such, could have done so on what was effectively coincidence, or chance occurrences. So,
    · I then took the new dataset, standardized it, ordered it according to frequency of occurrence (for OTU’s, i.e. most common / prevalent) and constructed new splines for all three datasets overlaid on the same plots. Pretty involved code, but I got there in the end. I have done this for the 100 most prevalent taxa (only, I will extend it to the top 200, but the number of useful OTU’s decreases with decreasing prevalence). See the 4 “Forams_RNA_AllSurveys_Top_X-X” figures. Black symbols for the initial (Year 1) MS dataset, blue is Year 2 MS and red is Year 2 BGB.
    · I have gone through each of these plots and assessed them according to how consistently the 3 independent datasets indicate ES range for each OTU, on the basis that if all three datasets say the same thing, then it is unlikely to be a coincidence, and therefore a robust means of assigning an EG. Basically, how close the 3 vertical lines are. I then allocated the EG’s to those OTU’s accordingly. This becomes my new Lookup list for the fAMBI.
    · So I now have EG’s allocated to specific OTU’s, as they are numbered in the new dataset – circumventing the transferability problem.
    · I then recalculated the fAMBI’s for the three datasets and compared. The three figures ending in “other plots” compare indicator stats other than ES and fAMBI. There are some trends, but some week relationships too.
    · The key plot is the “Forams_AllData_DNA_ES_v_fAMBI”. This is looking pretty good! Strong linear relationships for all three datasets. Now this is still a slightly contrived situation, as I have used all data to allocate EG’s and then applied it back to the same data. But the shear fact that the EG’s are allocated based on consensus between three independent datasets, gives me much more confidence in the result and hope that the next time it is tested against a new dataset, it will produce a good result (we need to do this at some point). Importantly, it does work for BGB!
    · I have made a few refinements to the way EG’s are categorized / defined (deviating slightly from Borja et al) to suit this process.
    · All in all I think there is the guts of a MS…
    · Now I would love to try integrating the bacteria results in a similar manner… As the relationships can only improve with refinement and the addition of new, good indicators. I still have only a few high level EG allocations (EG IV and V). Any chance that you could send that dataset through Soose so I can have a play?
    Now I reeeally need to do some other work for a bit. If there is any possibility that I can charge a bit of time somewhere for this, it would be appreciated because I have used 3+ days annual leave in addition to some IMR time so far. But understand your constraints. I might see if Chris has any ideas.
    Talk tonight.
    Nige
Introduction
The degradation of marine ecosystems is a global issue due to the many human-mediated impact pathways such as agriculture and horticultural discharges of fertilizers and stock effluent (Howarth et al. 2002, Smith et al. 2006), and point-source inputs of human and industrial wastes (Taylor et al. 1998, Bothner et al. 2002). An effect that is common to most of these stressors is the enrichment of the water column and seabed from additional nutrients. As a result, most environmental monitoring programs will include enrichment indicators, such as direct measures of nutrient concentrations (more common in the water column), concentrations and prevalence of nutrient responsive organisms (e.g. phytoplankton and benthic macrofauna), and several associated physico-chemical indicators relating to the oxygen status. The reliability and informative nature of these indicators is critical to our ability to detect change and status in relation to any real or perceived environmental thresholds.
Benthic macrofauna analysis is generally viewed as the ‘gold standard’ for assessing benthic condition, and with it, marine ecological quality status (), however, the method has two key limitations that are increasingly pertinent and therefore restrictive of its use. Importantly, the composition of the macrofauna community (i.e. the 1-20 mm sized animals that live within the sediments) provides a tangible, time-integrated picture of recent environmental influences by virtue of their presence and prevalence. Understanding of the relative sensitivities and the roles that each of the various macrofauna species play during disturbance and subsequent ecological succession reliably permits conclusions regarding the general level of ecological modification and disturbance. The limitations of this approach stem from the time it takes to sieve, sort, correctly identify and enumerate all of the individuals in a sediment sample. The labour costs of having skilled personal undertake this task in developed countries are significant, with analytical costs on the order of $100-1500 USD per sample (Author Pers. Obvs.). Such high costs can influence survey design, affecting both the frequency and intensity of sampling, and in doing so can adversely impact our ability to understand a system (temporally) and the reliability of the estimates (spatial replication). Moreover, the time consuming analytical process has the potential to create a major bottleneck in the work stream which can delay the provision of the results by several months, and with that our ability to report on and respond to the results in a timely manner.
This problem is not new, and much effort has gone in to developing other potential more cost effective indicators of enrichment, but as yet there is presently no universally accepted replacement. Biogeochemical indicators, such as redox potential ( Wildish) and total free sulphides (Hargrave) have gone some way to addressing this as they are both relatively cost effective and rapid, and often correlate reasonably well with biological indicators (Hargrave, Keeley, ??). However, they have one distinct limitation, in that they are one step removed from ecology and exceptions often exist where the correlations are poor, and as such they tend to be viewed as a compromised assessment, providing a proxy for ‘ecological health’. This may be in part related to differences in response times, and therefore indicator-specific time lags and the degree to which the chemical indicators are providing a temporally integrated picture of effects. Therefore there remains a clear and immediate need for an alternative method for evaluating ecological quality status that is both rapid and cost-effective, and directly related to ecology and biodiversity.
Another popular approach to evaluating benthic effects involves the application of biotic indices, which still require a full macrofauna assessment, and aim to condense what is effectively multivariate data in to a single number. There is presently a plethora of such indicators (for reviews see Pinto et al and ???) that use a variety of approaches, many of which are region and/or environment specific (e.g. MEDDOCC, mmmm). Some do this very effectively, have proven to be reasonably transferable and are becoming more broadly utilised as a result. One such group indicator utilise the Eco-Group approach developed by Borja et al. () for the application of the AZTI marine biotic index (AMBI). The basic premise here is that knowledge of the disturbance/enrichment tolerance of the dominant macorfauna taxa can be used in conjunction with their relative prevalence to calculate a scaled index. to composition and knowledge of species-specific sensitivities down in Some do it fairly well ().
Salmon farms as a case study - induce pronounced seabed enrichment
In this paper we… exploit the strong enrichment gradients that are commonly observed in association with salmon farms (Ref, Keeley) to test the potential for using NGS of foraminiferal communities to determine benthic ecological quality status. To do this we use the overall synthesis indicated by multiple established indicators of benthic enrichment to elucidate the ecological characteristics of lesser known foramifera taxa and then to allocate Eco-Groups (Borja et al) in a manner analogous to established macrofauna methods (see Keeley et al. 2012). The Eco-Group allocations are then used to derive and test the suitability of a variety of established indicators for purposes of discerning benthic enrichment stage.