Mangrove Zonation as a Tool to Infer the Freshwater Inflow Regime in the Data-Poor Ruvu Estuary, Tanzania

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

MS title: Mangrove zonation as a tool to infer freshwater inflow regime in the data-poor Ruvu Estuary, Tanzania.

The study addresses an important management and conservation issue using a creative ecological proxy. However, the current version lacks sufficient analytical rigor and methodological transparency to support its broader applications. Enhancing quantitative depth, methodological details and modeling clarity will substantially improve its scientific and practical impact.

Below are specific comments

-The abstract lacks specific quantitative findings for example, measured salinity ranges, identified zonation distances, and model parameters.

- Please add key numerical results to the abstract and specify the period and duration of field data collection.

-In the introduction, explain how mangrove zonation quantitatively correlates with salinity to provide scientific justification for the approach.

- Include a clear statement of hypotheses or research questions at the end of the introduction.

Materials and methods: The methods rely heavily on qualitative and visual observations. Quantitative measures like vegetation density, canopy cover, species dominance indices are missing.

- Add correlations between salinity and species presence/abundance to support the interpretations.

- Provide justification for conducting the survey during a single seasonal window (June 18–27, 2013).

- Clarify the calibration and validation procedures for salinity measurements

- Were replicate transects or repeated measurements conducted to assess spatial/temporal variability?

-How were the vegetation survey sites chosen, randomly, systematically, or based on accessibility?

-Add sediment characteristics like grain size, organic matter etc.

- The results mix descriptive observations and interpretation; please separate these.

- Provide statistical summaries (means, standard deviations, regression analysis) of salinity data.

- The discussion relies heavily on qualitative explanations; Add more quantitative analysis.

- What is the estimated error margin for salinity readings at different depths?

- The modeling approach is only conceptually discussed, it is important to present and validate the model.

- Discuss the potential for integrating remote sensing and long-term monitoring more explicitly with GIS data.

- Was any numerical modeling performed, or is the modeling discussion hypothetical?

Author Response

REVIEWER 1 - Comments and Suggestions for Authors

MS title: Mangrove zonation as a tool to infer freshwater inflow regime in the data-poor Ruvu Estuary, Tanzania.

 

The study addresses an important management and conservation issue using a creative ecological proxy. However, the current version lacks sufficient analytical rigor and methodological transparency to support its broader applications. Enhancing quantitative depth, methodological details and modeling clarity will substantially improve its scientific and practical impact.

Firstly, we sincerely thank you for taking the time to carry out a thorough and detailed review and for your suggestions, most of which we have considered, and that have much improved the clarity, completeness and organization of the paper. We have included more details on methodology, so that the study can be replicated by anyone. The method is based on identifying mangrove species from high resolution boat-based photography, as well as taking salinity measurements at various tide conditions daily for two weeks. As you also mention, mangrove zonation is used as a proxy for the actual salinity regime. This study has not gone further into modeling. However, as there is a vast literature on salinity estuarine modeling, we just provide some references of studies to indicate how the salinity regime inferred from mangrove zonation could inform the models.

Given the urgency in figuring this out for informed management of estuaries, and by connection, water resources throughout the basin, our intention is to suggest the use of mangrove zonation as an easily achievable method to determine the salinity regime, in the absence of long term freshwater inflow, tidal flow and salinity data that would otherwise be required to estimate freshwater inflows needed to maintain estuarine ecosystems. This data is simply not there in much of the underrepresented world – Africa, Asia, Latin America / Caribbean and Oceania. Hence, our intention is that the paper provide a useful resource for water resource managers and planners, and hence is not a classic scientific study with hypotheses and statistical data analysis.

 

Below are specific comments

 

-The abstract lacks specific quantitative findings for example, measured salinity ranges, identified zonation distances, and model parameters.

thank you for these suggestions; we have added the measured salinity ranges, zonation distances. We did not actually do any modeling.

 

 

- Please add key numerical results to the abstract and specify the period and duration of field data collection.

this has been now done

 

-In the introduction, explain how mangrove zonation quantitatively correlates with salinity to provide scientific justification for the approach.

getting a “quantitative” correlation is perhaps difficult because salinity changes continually with tide and season, so one has to use the average salinity regime. But in Ruvu estuary (and thousands of other estuaries worldwide) there is no long-term salinity data available. Hence we use the proxy approach of individual mangrove species salinity tolerance ranges, which has been established in the literature on mangroves in this part of the world. We mention that along with references in the section on mangrove zonation in the introduction, and later in results.

 

- Include a clear statement of hypotheses or research questions at the end of the introduction.

done, thank you we agree that makes the paper clearer and easier to grasp.

 

Materials and methods: The methods rely heavily on qualitative and visual observations. Quantitative measures like vegetation density, canopy cover, species dominance indices are missing.

the major part of the survey was boat-based because 1. we wanted to see what trees occur on the riverbank, because these trees are in direct contact with river water, and hence would be the best indicators of what salinity conditions have been prevailing in the estuary over their life span. 2. being a long estuary - 28 km, boat based survey is practical, 3. baring a few places with beaches, it is hard to access the shore  as the trees overhang the water. Treacherous slippery roots/branches, encrusted with sharp mussels and barnacles, and presence of Nile crocodiles and hippos.

So we took photos every 15 seconds - we now include this in the methods for better clarity. The mangrove species being just 8 are very easy to identify from photos - its a time consuming exercise as there are thousands of photos, but certainly doable with minimal training.

We did not measure species density, canopy cover and species dominance indices, as our aim was to just to identify species occurrences along the riverbank. Species density and the other variables would be useful in a periodic monitoring study that seeks to look at change in forest composition over time in response to various abiotic and biotic factors. However thank you for suggesting this, we will mention this in the methods section.

We did have three shore-based transects, to verify if higher ground inland had mangrove spp with lower salinity tolerance than those on the riverbanks. However we did not include that data here as this was done only at one place where we could access the forest via a sandy beach. Instead, we just mentioned the broad observations in section 3.2.3 ( Elevated sandy banks).

- Add correlations between salinity and species presence/abundance to support the interpretations.

figure 5 illustrates that - relating species composition categorized into vegetation types, with the salinity data from our 2 week study, at various tide conditions. However, as we did not have salinity data over a year, we could not correlate species occurrences with the average salinity regime prevailing in the estuary.

However, a correlation with salinity data would not provide a completely accurate picture, as we do not have salinity data over a year - in wet seasons with higher freshwater inflows, one would expect to get low salinity throughout the estuary, while in dry season, the salinity values would be higher, stemming from increased seawater intrusion into the estuary enabled from reduced freshwater inflows. Furthermore there can be interannual changes in freshwater inflows depending on precipitation and land use change in the basin, which can ultimately result in interannual changes in salinity regimes in the estuary, that in turn affects the riparian vegetation.

Besides, average values are a longterm abstraction of what a plant faces, but in terms of tolerance, it’s the maximum salinity encountered in the root zone as well as the duration that affects the plant

Additionally, the salinity ranges of these mangrove species has been well documented in the literature * references provided. All we did was to observe the species composition along the 28 km on both sides of the river, group that into vegetation classes and intensively take salinity data every day, several times a day going up and down the river under diff tidal conditions, to see the salinity conditions in the seasonal transition, to avoid both seasonal extremes. 

 

- Provide justification for conducting the survey during a single seasonal window (June 18–27, 2013).

done in manuscript. “Financial and technical logistics permitted only one 2-week long expedition with several trips a day at both flood and ebb tides along the entire 28 km upstream end-estuary mouth distance. Hence the transition season was chosen, to observe and record water salinity between the expected predominantly fresh conditions at peak wet season high flows and saltiest conditions in the dry season, when the freshwater inflows are at their annual minimum.

“

- Clarify the calibration and validation procedures for salinity measurements

now done

 

- Were replicate transects or repeated measurements conducted to assess spatial/temporal variability?

10 replicates (10 days) of boat-based salinity measurements along estuary length under various tidal conditions. 3 replicate transects for shore-based survey of mangrove species comp with distance from riverbank.

 

-How were the vegetation survey sites chosen, randomly, systematically, or based on accessibility?

 riverbank vegetation survey was done continually as the boat moved slowly (~5 mph ) with frequent stops every 100 m for water quality measurements. Photographs were taken on both sides of the riverbank. Hence almost the entire riverbank was observed and recorded. For the onshore transects that stated at the river bank and went inwards perpendicular to the river bank, these were taken at a place where the boat was able to dock - small sandy beach.

 

-Add sediment characteristics like grain size, organic matter etc.

we did not measure these variables, as these fell outside the scope of the study. If we were studying plant physiology of the mangroves in terms of water uptake and soil salinity then these would be useful. But we are just looking at what mangroves exist, to get a proxy of the average salinity regime, and not the mechanisms allowing them to exist where they do.

 

- The results mix descriptive observations and interpretation; please separate these.

done. That being said, we have retained some discussion in the results for mangrove plant community zones, as its easier to discuss these mangrove occurrences with references in one place, rather than splitting it up.

 

- Provide statistical summaries (means, standard deviations, regression analysis) of salinity data.

 Given that we collected salinity data just for two weeks, that too under constantly changing tidal conditions during the day (and not night as it was manually connected, not from an automated setup), we felt that statistical description of the data would not have much relevant meaning. We have, however, included the raw data as supplementary data, for readers to use if they are interested.

 

- The discussion relies heavily on qualitative explanations; Add more quantitative analysis.

as explained earlier, there is limited scope for quantitative analysis here. This study relied upon high resolution photos taken every 15 seconds, thus covering almost the entire estuary riverbank, as well as salinity data over 2 weeks. The purpose is to demonstrate this approach as a practical easy to apply tool towards better management of estuarine ecosystems ultimately.

 

- What is the estimated error margin for salinity readings at different depths?

has been included in methods now; “The resolution of the Ecosense 300A is 0.1ppt with a stated accuracy of ± 0.2% and a measurement range of 0-70 ppt.”

 

- The modeling approach is only conceptually discussed, it is important to present and validate the model.

the aim of this study was to see if a mangrove composition survey can provide information for freshwater inflow-seawater mixing models for estuaries, which we refer to. There is a vast literature on various generations of models, out of which we referred to some that were applied in estuaries with minimal data. We did not proceed with modeling, as that was out of scope of our study period and resources. Several hydrologists from the Ministry of Water in Tanzania, and Sokoine Agricultural University expressed interest in modeling with data from this study; however that is out of scope of the current paper.

 

- Discuss the potential for integrating remote sensing and long-term monitoring more explicitly with GIS data.

 Thank you for suggesting this – we have included some more info on GIS in the Discussion ( Section 4.2), pointing out layering species occurrences on a DEM, and the challenges of obtaining a DEM of sufficient vertical resolution – for instance, the current Landsat collection 2 that’s freely available has a vertical resolution of 30 m – useless when a one meter elevation difference can completely change mangrove composition. LIDAR is better, can even get 1-2 cm vertical resolution, however its expensive as it has to be generated using specialized cameras on drones or low flying planes. Another challenge is change in elevation resulting from either land subsidence/tidal washout or sediment accretion, that can require updated LIDAR images every couple decades. However, it is still worth obtaining a LIDAR image, in an effort to ultimately maintain estuarine ecosystem services that are worth much more.

- Was any numerical modeling performed, or is the modeling discussion hypothetical?

as mentioned before, no modeling was performed. In the paper we mention that models require inflows and salinity concentrations, which are not available in most estuaries in Asia/Africa/Latin America. Hence the mangrove zonation data can provide clues as to the average salinity regime in different parts of the estuary. And that inferred salinity regime can then inform any estuary model. We have provided a few references for such models used in the literature, as a starter. However selecting models and proceeding to parameterize and calibrate+validate them is a whole different endeavor.

Reviewer 2 Report

Comments and Suggestions for Authors

MS presents very interesting research, which has great importance in the view of biodiversity conservation and Climate change. Also this research has a great importance for understanding of the biodiversity and environmental conservation in Africa.  However MS has some shortcomings which should be corrected.

First of all - in the data is Water, 2023, 15, however now 2025.

Introduction

It is too long. As a rule, Introduction should contain common brief representation of the studied scientific question, task of the study or null - hypothesis. In MS Introduction should be devided as minimum on two sections - Introduction as well and Study area. 

Figure 2 - is this part of results or someone published data? Please clarify this and, if necessary transfer these data in result section.

 Authors mention some report GLOWS 2015, but dont clarify which data of this report were included into MS. Also i havent found information about season, year and replications of the collected data. it should be given in material and methods section. When authors collected the data, does they count number of plant individuals and their density? Could it be correlated with the level of salinity? 

Also for me, this is not clear if authors used the satellite data? 

The section of results contains mix of results and discussion. This makes this chapter difficult for understanding. Please, divide this part on two: Results and Discussion. In Results the data should be represented and in Discussion - all information about their correspondence to ther surveys, possible reasons and explanations etc.

In Result section - every mention of the relationship between fact (character of vegetation) and factor  (salinity) must be accompanied by specific numbers (changes in density or numbers of individuals etc) or calculated correlations.

As conclusion/ my opinion is that this MS cannot be published at present state and needs serious revision

 

Author Response

REVIEWER 2 - Comments and Suggestions for Authors

MS presents very interesting research, which has great importance in the view of biodiversity conservation and Climate change. Also this research has a great importance for understanding of the biodiversity and environmental conservation in Africa.  However MS has some shortcomings which should be corrected.

We sincerely thank you for your time to carry out this detailed review and for your suggestions, most of which we have considered, and that have much improved the clarity, completeness and organization of the paper.

First of all - in the data is Water, 2023, 15, however now 2025.

Not sure if we understand this comment – maybe the Special Issue opened in 2023 ?

 

Introduction

It is too long. As a rule, Introduction should contain common brief representation of the studied scientific question, task of the study or null - hypothesis. In MS Introduction should be devided as minimum on two sections - Introduction as well and Study area.

thank you for this suggestion - we agree and have now made it more concise, and restricted to the main topic of this paper, ie the use of mangrove zonation to determine average salinity regime. We shifted the section on vulnerability of the Ruvu estuary to basin-wide water management, climate and land use change to the Discussion - we feel this is important to include the basin-wide perspective, especially as our intention is that this paper can benefit water managers an policy makers for actual implementation of sound estuary management.

We agree that what you describe applies to a paper describing a specific classic scientific experiment. Our paper however has several objectives, the foremost being how to quickly use mangrove zonation as a proxy for long-term interdecadal salinity data that is SIMPLY not available, and in much of the world will continue to be unavailable, owing to the high cost, technical complexity  and field challenges ( such as vandalism, wildlife damage) to install and operate monitoring networks. It thus describes a methodological approach, in a situation with very little time available to manage fast degrading estuaries under increased pressure. In that sense, it is more of an environmental management paper than one looking at a single specific question with null and alternate hypotheses and statistical testing.

Figure 2 - is this part of results or someone published data? Please clarify this and, if necessary transfer these data in result section.

This figure has been made by the authors with available long-term river discharge data from the Govt of Tanzania. It is also included in technical reports written by the same authors, but has not been published in any scientific peer-reviewed publication. This graphic portrays the flow regimes in different tributaries across the Ruvu Basin, which collectively influence the freshwater inflows arriving in the estuary. As there is no discharge data in the estuary, with the closest station being 45 km upstream from the estuary, this data is useful to understand the seasonality of freshwater inflows into the estuary. For instance, based on these data, it can be expected that the peak flows in the basin happen between Aprul and June, hence the highest freshwater flows would occur in May-July into the estuart. Likewise the lowest flows are in Auig-Oct, and hence that period would be the most saline conditions in the estuary. Our intention and hope is that water resource managers and policy makers can find this publication of practical use, not only for the Ruvu basin, but estuaries elsewhere in the underrepresented world.

(Note now this is Figure 8 as we have shifted that section to the Discussion, following your suggestion of making the Introduction more concise and focused on mangrove zonation – the main part of the paper).

 Authors mention some report GLOWS 2015, but dont clarify which data of this report were included into MS. Also i havent found information about season, year and replications of the collected data. it should be given in material and methods section. When authors collected the data, does they count number of plant individuals and their density? Could it be correlated with the level of salinity?

We have now added the GLOWS2015 reference information in several places in the paper, including Figure 8 (which was Fig 2 ). The methods section now has additional data, allowing it to be replicated by readers.

Re: survey, we did not count number of individuals nor their density, as our aim was to just to identify species occurrences along the riverbank. Species density and the other variables would be useful in a periodic monitoring study that seeks to look at change in forest composition over time in response to various abiotic and biotic factors. However thank you for suggesting this, we will mention this in the methods section the reason why we did not include species abundance and density.

We did have three shore-based transects, to verify if higher ground inland had mangrove spp with lower salinity tolerance than those on the riverbanks. However we did not include that data here as this was done only at one place where we could access the forest via a sandy beach. Instead, we just mentioned the broad observations in section 3.2.3 (Elevated sandy banks).

Because our aim was just to record the species occurring along the riverbank, the vegetation survey was done continually as the boat moved slowly (~5 mph ) with frequent stops every 100 m for water quality measurements. Photographs with GPS coordinates were taken on both sides of the riverbank, at first every 15 seconds, and later on every minute. Hence almost the entire riverbank was observed and recorded. The thousands of photographs were then used to note down species occurring , a subset of which was then added as points in Google Earth, to denote monospecific stands, and milti-species occurrences. Prior to importing into ArcGIS as a shapefile, these points were categorized into 4 groups as

(i)mangroves that are fully tolerant of marine salinity levels (30-35 ppt) and long periods of continuous inundation

(ii) mangrove species with intermediate tolerance to salinity level (brackish conditions: 10-30 ppt)

(iii)mangroves and palms with low tolerance to salinity (< 10 ppt), and

(iv)plants intolerant of salinity.

It is hard to correlate with salinity, since salinity at any point is constantly changing with the tide and the season – wet season with high freshwater flows make it low, while in dry season, the salinity gets higher. We had data just for two weeks. Taking an average of that would be incorrect, as what the plants experience is the actual salinity (and duration of exposure) and not the average. Species salinity tolerances for Tanzania are also documented in the literature. So we just mapped the species composition, and layered it on our salinity data  (at the surface at high tide), to visually see the relationship.

Also for me, this is not clear if authors used the satellite data?

No, we did not. We just mentioned that an important application of the mangrove composition data obtained by surveys like ours is the calibration/validation of satellite-sourced images, that are invaluable in monitoring wide areas.

 

The section of results contains mix of results and discussion. This makes this chapter difficult for understanding. Please, divide this part on two: Results and Discussion. In Results the data should be represented and in Discussion - all information about their correspondence to ther surveys, possible reasons and explanations etc.

Thank you. We have now separated these. We still retained some discussion with the results describing the various plant communities along the estuary, as it would be easier for the reader to read inferences and references in one place.

 

In Result section - every mention of the relationship between fact (character of vegetation) and factor  (salinity) must be accompanied by specific numbers (changes in density or numbers of individuals etc) or calculated correlations.

As explained earlier, it is difficult to quantitatively relate salinity to species occurrence, because salinity at any point in the estuary is continually changing twice a day ( tides ) and seasonally (variable freshwater inflows). One could take average salinity over a year, however we do not have data for a year, and besides, I do not know how useful it would be to correlate average salinity with species occurrence, as the plants do not experience average salinity over the year – it’s the high salinity values and period of exposure that determine salt stress in a plant. However, a map like Fig 5 can visually portray the salinity measured at certain tidal condition and time of year to species composition.

As conclusion/ my opinion is that this MS cannot be published at present state and needs serious revision

We hope the manuscript has improved after taking into account the majority of your suggestions, and find it acceptable for publication, and readership not only by academic scientists but by water resource managers, basin planners, policymakers and all who are concerned with protecting and maintaining estuaries and their ecosystem services.

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

No more comments. The manuscript now meets the standards expected for publication in

 Water.

Reviewer 2 Report

Comments and Suggestions for Authors

The authors made significant changes in MS. This work is very valuable in the view of the nature conservation and represents new? previously unpublished data on the hydrological conditions and distribution of mangroves in Ruvu Estuary. The authors made corrections according to all my comments and this MS can be accepted for publication