The Southeast Alaska Tribal Ocean Research (SEATOR) Partnership: Addressing Data Gaps in Harmful Algal Bloom Monitoring and Shellfish Safety in Southeast Alaska

Round 1

Reviewer 1 Report

One of the biggest barriers to HAB prediction, is the multitude of monitoring data that is needed to produce robust models.  This paper presents PST data from shellfish samples across a traditionally under sampled region.  

This paper is not novel in that monitoring for HABs has been done, but it is NEEDED because monitoring for HABs has not been done everywhere and local variances in HAB dynamics are bound to occur.  

There are only a few minor changes I would make to the manuscript in order to make it easier to understand. 

1. Since the Materials and Methods are at the end of the paper, the abbreviation for STAERL is defined after it is used in the text.  STAERL is first used as an abbreviation in line 146 but not defined until line 221.
2. "Microalgae" is used in the abstract, however "phytoplankton" is used throughout the main manuscript.  Since microalgae is only mentioned the one time in the abstract (line 7), it should be changed to phytoplankton to be more consistent with the remainder of the text.
3. Scientific names are missing italics throughout the text.  The Materials and Methods section (4) is the only one in which the scientific names are correctly and consistently italicized.  Issues exist throughout the text and captions for the abstract and sections 1-3.  
4. Section 2.1 appears to begin with a justification as to why refractometer measurements were used for salinity instead of using sensor data.  However this was not clear until the Materials and Methods section was read, at the end of the manuscript.  A little clarification in the first few sentences to highlight the fact that hand grabbed samples were used in salinity determinations is needed.  It is quite clear that sensors were not used and why, it is not clearly stated what was used instead.  
5. How was the ordinal scale determined for relative abundance?  What were the thresholds?  The graph with the Alexandrium relative abundance data shows a max of 0.2, is this present? or common? or bloom?  Many people use net tow relative abundance data in their monitoring projects, but every lab has different thresholds.  A little more detail is needed so that people may make comparisons to their own data a little bit easier.

Author Response

"One of the biggest barriers to HAB prediction, is the multitude of monitoring data that is needed to produce robust models.  This paper presents PST data from shellfish samples across a traditionally under sampled region.  

This paper is not novel in that monitoring for HABs has been done, but it is NEEDED because monitoring for HABs has not been done everywhere and local variances in HAB dynamics are bound to occur."

We thank the reviewer for their comments and we agree this monitoring program is needed and valuable.

"There are only a few minor changes I would make to the manuscript in order to make it easier to understand. 

Since the Materials and Methods are at the end of the paper, the abbreviation for STAERL is defined after it is used in the text.  STAERL is first used as an abbreviation in line 146 but not defined until line 221."

This has been corrected (Line 106).

"Microalgae" is used in the abstract, however "phytoplankton" is used throughout the main manuscript.  Since microalgae is only mentioned the one time in the abstract (line 7), it should be changed to phytoplankton to be more consistent with the remainder of the text."

We thank the reviewer for their attention to detail, this has been changed. Note that since the journal submission process has us upload the abstract as plain text it will not show up as a tracked changes in the word document, but it was changed in the submission.

"Scientific names are missing italics throughout the text.  The Materials and Methods section (4) is the only one in which the scientific names are correctly and consistently italicized.  Issues exist throughout the text and captions for the abstract and sections 1-3."

We thank the author for this review. We aren't sure why the italics didn't appear in the submission but it was a mistake in formatting the submission for this journal. We have edited the scientific names appropriately throughout the manuscript. 

"Section 2.1 appears to begin with a justification as to why refractometer measurements were used for salinity instead of using sensor data.  However this was not clear until the Materials and Methods section was read, at the end of the manuscript.  A little clarification in the first few sentences to highlight the fact that hand grabbed samples were used in salinity determinations is needed.  It is quite clear that sensors were not used and why, it is not clearly stated what was used instead."

We agree that having the methods below the results has resulted in some confusing text in the results, but we have made the edits the reviewer suggested in order to make this section more clear (beginning line 68).

"How was the ordinal scale determined for relative abundance?  What were the thresholds? 

We have added text in the methods to explain the relative abundance metric, and where it was adapted from (line 228). 

The graph with the Alexandrium relative abundance data shows a max of 0.2, is this present? or common? or bloom?  Many people use net tow relative abundance data in their monitoring projects, but every lab has different thresholds.  A little more detail is needed so that people may make comparisons to their own data a little bit easier."

To answer this question, the graph the reviewer is referring to Figure 4c is the proportion of samples with Alexandrium (present, common or bloom). So a proportion of 0.2 indicates that 20% of the samples from the month of June contained Alexandrium, or a value of 0 would indicate that no samples from that month would contain Alexandrium. This is related to the ordinal scale insofar as the groups of present, common and bloom were lumped together to form a simple presence/absence value. We have updated the figure caption to make this more clear (Figure 4). 

Reviewer 2 Report

Author Response

"Since 2016, SEATOR has analyzed the relationship between these and shellfish toxins using 2,400 field observations of salinity and sea water temperature and 1,700 bivalve samples. It is expected to contribute to the rapid delivery of important information to neighboring residents as well as to humans who eat shellfish worldwide. According to the author's explanation, it may or may not have seasonality, but it is a report that obtains good results that changes in seawater and water temperature have a high relationship with the concentration of toxins produced by shellfish. "

As stated in the manuscript, some species exhibit a high degree of seasonality with respect to PST concentrations (i.e. mussels) while other species do not exhibit a strong seasonal pattern (butter clams). The seasonality is of course is intimately linked to the seasonal cycle of temperature which drives primary production. The reviewer does not appear to be suggesting any edits here, merely providing a summary of the manuscript.

"However, as the authors say, if shellfish produce toxins as a defense, research on the kind and density of predators should be accompanied."

The SEATOR phytoplankton monitoring program was developed and guided by the nationwide Phytoplankton Monitoring Network (PMN). This program has developed a user friendly and broadly executable protocol for sample acquisition and analysis that does not require noting or quantifying the number of predators and grazers. While we recognize the value that the addition of grazer abundance would add, these data don't exist and to incorporate them into the this and other monitoring efforts would necessitate major revisions to a broadly adopted protocol. We have added a statement in the conclusion about collecting predator data and hope that future efforts will secure funding to do this research, but the SEATOR network has not and does not plan to collect predator data going forward. 

"In particular, inflow of fresh water is also expected to play a very important role."

We agree with the reviewer that freshwater input plays a role in the development of marine HABs. The goal of this paper was to describe the methods and a summary of the data collected by the SEATOR network and not to develop sophisticated models of HAB dynamics in Southeast Alaska. There are groups that monitor freshwater input in this region (i.e. USGS) who's data are publicly available, and we encourage the incorporation of these data into sophisticated models (we noted this in section 2.3), however that would be beyond the scope of this paper. 

"In particular, many shellfish toxin experts report that the kind and concentration of PST is determined from the chemical ratio of nutrients or the concentration of specific nutrients. In this or later study, the study on the concentration of nutrients in the waters corresponding to the kind and concentration of toxins must be explained."

Similar to the two above suggestions, the SEATOR program we have described here does not collect samples for nutrients as part of their regular sampling. We recognize the value of nutrient data in HAB formation and toxin production, but we emphasize that while the SEATOR network does conduct and support research their overarching goal is to provide relevant and timely data to harvesters and communities. Nutrient data, while informative, is not easily analyzed in real time and is beyond the scope of the SEATOR program in and of itself. 

We noted in the conclusion that future efforts assessing the role of nutrient concentrations are warranted and could benefit from the SEATOR dataset. 

We also emphasize that while in some regions nutrient influx via riverine discharge and cultural eutrophication might be a strong driver of HABs in recent decades, Southeast Alaska has had natural HABs of PST producing phytoplankton for hundreds of years. Nutrient concentrations certainly play a role in Alexandrium dynamics in Southeast Alaska, but are likely not as integral a driver as they might be in other regions such as the Gulf of Mexico or the West Coast. 

"In conclusion, it is judged that it is difficult to continuously provide good information to people related to shellfish production and toxin management by using this type of research or monitoring alone."

While we agree that it is difficult to continuously provide shellfish toxin and HAB abundance data, we feel that the SEATOR network has established protocols and funding streams that allows the network to do exactly that. We disagree with the notion that because these data do not describe other aspects of HAB or shellfish research that they are useless or uninformative on their own. While other states and regions might have longer, more frequent, or more involved sampling of HABs and shellfish this is the first time that a regional multi-year monitoring program has been created in Southeast Alaska. Perhaps more importantly, this program is relatively unique in its partnerships and collaborations between Tribal entities, Universities, and federal agencies and can be a valuable template for other regions both within Alaska and across the world.