the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 10 Aug 2026
Stream connectivity and governing processes in high-mountain wetlands
Abstract. High-mountain wetlands (hereafter wetlands) have the potential to store and release large volumes of water. They provide substantial contributions to water supply for highland communities and receiving lowlands and their extended residence times can buffer water demand during drier periods. Despite their hydrological significance, major gaps remain in understanding the connection between wetlands and streams, which prevents us from accurately conceptualising them in models. We used a combination of conservative fluorescent tracing and monitoring of rainfall-runoff and wetland water levels to develop a perceptual model of connectivity between wetlands and streams. The experiments were conducted during the wet or dry season in five study catchments (0.38 km2 – 12.58 km2) from northern Ecuador to southern Peru. Fluorescein was introduced into wetlands and monitored downstream with activated carbon samplers for 5–12 months. Our results suggest travel times from less than 1 week to upwards of 3 ½ months, with one experiment seeing little to no response. Results indicate that wetlands have a stronger hydrological connection to streams in the wet season than in the dry season, where in some cases we observed no connection at all. Multiple peaks in fluorescein concentration from a single injection during the wet season may suggest that the wetlands contribute to streamflow via multiple pathways and processes. The results demonstrate a complex connection between wetlands and streams, controlled by topographic configuration and season, amongst other factors. However, persistent contributions from wetlands to streams observed several months after dye introduction support their significance to downstream, year-round water supply. The results provide improved understanding of wetland-stream connectivity that informs the development of models to address the need for site-specific evidence of wetlands’ influence on water supply and the ability to scale this information in space and time.
Competing interests: At least one of the (co-)authors is a member of the editorial board of Hydrology and Earth System Sciences.
Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this paper. While Copernicus Publications makes every effort to include appropriate place names, the final responsibility lies with the authors. Views expressed in the text are those of the authors and do not necessarily reflect the views of the publisher.- Preprint
(3112 KB) - Metadata XML
-
Supplement
(15 KB) - BibTeX
- EndNote
Status: final response (author comments only)
- RC1: 'Comment on egusphere-2026-4439', Anonymous Referee #1, 29 Sep 2026
-
RC2: 'Comment on egusphere-2026-4439', Anonymous Referee #2, 04 Oct 2026
The paper Stream connectivity and governing processes in high-mountain wetlands by Ross et al. aims to assess hydrological connectivity between wetlands and streams in high-Andean ecosystems, namely the páramo and Puna regions of Ecuador and Peru, using a combined approach of fluorescent tracer injections and hydrometeorological data collected at five sites across the Andes. Although the research topic is highly relevant to the region, the manuscript presents several important methodological and conceptual shortcomings that, in my view, currently preclude publication. I outline the major concerns below.
1. Experimental setup
The experimental setup is concerning in both spatial and temporal dimensions. As can be seen from the different landscapes presented in Figure 1 and the associated publications describing the different sites, the monitored wetlands occur under substantially different geological, geomorphological, biological, and ecological conditions. These include differences in groundwater connectivity, glacier and snowmelt contributions, soil organic matter content, hydrophysical properties, and parent material or geological origin (e.g., volcanic versus non-volcanic substrates).
Although this diversity could potentially be considered a strength of the study, the monitoring scheme shown in Figure 1 does not demonstrate that a consistent set of criteria was used to make the results across the different study sites comparable. For example, Table 1 suggests that the selection of tracer injection locations did not follow a common framework among catchments based on factors such as wetland size, topographic gradient within the wetland and catchment, distance from the wetland to the stream, and the presence or absence of glaciers among others. Similarly, the locations at which tracer samples were collected do not appear to have been selected according to a standardized protocol that would ensure robust comparability among study areas.
The temporal component of the experimental design presents a similar issue. There does not appear to be a unified criterion for selecting the antecedent hydrological conditions, rainfall-runoff conditions, or a common framework for tracer injections during dry and/or wet periods.
This is a particularly important consideration when artificial tracers are injected into hydrological systems because system responses are strongly dependent on the hydrometeorological conditions prevailing at the time of tracer injection. Nevertheless, the dates of tracer injection at each site appear to have been selected independently, which substantially limits the validity and representativeness of comparisons among sites and wetness conditions.
It is also worth noting that, although tracers were injected during both wet and dry seasons in some catchments, other catchments were assessed during only one season. Again, hydrometeorological conditions at the time of tracer injection can have a major influence on the activation or deactivation of different flow paths. Consequently, without a common framework for the spatial and temporal selection of injection and stream sampling sites, the experimental design does not provide a sufficiently robust basis for interpreting hydrological processes.
2. Background and context
In the Introduction, the authors state that the use of natural tracers has been important for improving the understanding of high-Andean hydrology. A substantial portion of the section is devoted to discussing transit times of stream water and its connection with wetland areas based on natural tracers. This gives the reader the impression that the manuscript will focus primarily on natural tracers, such as isotopes and electrical conductivity, and on transit-time estimation, rather than on travel times estimated through artificial tracer injections.
This is an important weakness in the framing of the study because there is insufficient discussion of the use of artificial tracer injections to understand wetland–stream connectivity in other environments. Indeed, only two studies using artificial tracers to assess groundwater–surface-water connectivity are mentioned, and these studies were conducted in substantially different environmental settings. They are therefore used only briefly as background to a methodology that constitutes the central component of the present study.
The authors should provide a more comprehensive state-of-the-art review of the use of artificial tracers, particularly fluorescent tracers, in hydrological studies. This is especially important for systems containing substantial amounts of organic-rich material, where interactions between the tracer and the porous medium may affect tracer retention and recovery.
3. Specific characteristics of the wetlands
Although a general characterization of the study sites is provided, the specific properties of the wetland materials should have been characterized in considerably greater detail. Relevant properties include texture, porosity, organic matter content, chemical composition, as well as the depth and type of material into which the tracers were injected.
These characteristics should have been systematically documented across the monitored wetlands so that their influence on tracer retention and transport could be evaluated. Without this information, interpreting the observed tracer responses in terms of underlying hydrological processes becomes highly challenging.
This information is particularly important given the substantial variability in climatic, geological, and ecological conditions across the tropical Andes and the corresponding variability in the characteristics of the organic material accumulated within these wetlands.
4. Use of artificial tracers in highly organic systems
This is a particularly important issue that is not adequately addressed in the manuscript. The use of artificial tracers in environments characterized by high organic matter contents requires careful consideration because of the potentially high tracer-retention capacity of organic-rich media. Such retention can affect tracer recovery and consequently influence the estimation of subsurface travel times.
Indeed, tracer studies in organic-rich systems have considered the interactions between tracers and organic material before using tracer responses to interpret hydrological and ecological processes. This consideration is particularly important in the present study because the climatic and geological conditions differ substantially across the tropical Andes, and consequently the characteristics of the organic material accumulated in the wetlands are likely to vary considerably.
Therefore, an assessment of how different organic-matter conditions may affect tracer retention and transport should be conducted before drawing hydrological conclusions from the tracer experiments. At a minimum, the authors should provide a detailed discussion of this potential source of uncertainty and demonstrate that differences in tracer recovery among sites are attributable to hydrological connectivity rather than to differences in tracer retention by the wetland material.
5. Paper structure and organization
The structure of the manuscript is rather unconventional. The main body of the paper contains only one figure presenting the tracer-injection results, while much of the remaining information is relegated to the appendix. Furthermore, the Results and Discussion sections refer to Figure 2 only a few times. The remaining tracer experiments are presented primarily in the appendix and are not systematically incorporated into the discussion.
This substantially weakens the arguments presented by the authors and makes it difficult for the reader to evaluate how the individual tracer experiments support the conclusions.
Given the number of experiments conducted, I would have expected the authors to present and describe the results of each tracer injection in greater detail in the main body of the manuscript and then discuss the individual experiments and their similarities and differences systematically. The appendix should provide supplementary information rather than contain a substantial portion of the evidence on which the main conclusions are based.
6. Criticism of other methods
In the Introduction, the authors argue that robust and easily reproducible methods are needed to assess connectivity mechanisms across multiple sites and to quantify their physical and climatic controls.
However, based on the methodological issues associated with tracer injection and monitoring in streams, together with the limited consideration of the influence of high organic matter contents on tracer behavior, I question whether the approach presented here can currently be considered a robust and readily reproducible methodology for hydrological studies in high-Andean wetlands.
The manuscript also implies that the proposed approach offers advantages in terms of applicability and reproducibility. However, the laboratory analyses required are highly specialized and may not be readily accessible to local communities, water managers, or other stakeholders, both in terms of analytical capacity and the processing and interpretation of the resulting data.
Other approaches commonly used in high-Andean hydrology—including stable isotopes, geochemical tracers, and electrical conductivity—also have their own limitations and advantages, but have provided substantial information for developing process-based understanding of hydrological behavior across scales through well-established methods. Therefore, the manuscript should reconsider its claims regarding the ease of application and reproducibility of the proposed method in comparison with other tracer-based approaches.
Indeed, based on the information presented in the appendix for the different tracer injections conducted at different sites and during different seasons, several observations appear to be open to alternative interpretations and might even result from issues related to data collection. This raises questions about whether the proposed approach can genuinely be considered an easily reproducible and robust methodology without a more systematic framework for interpreting the tracer responses.
7. Generalized conceptual model
This is one of the most problematic aspects of the manuscript. Given the authors' stated objective of addressing a highly relevant issue in high-Andean hydrology across a large geographical region, it is surprising that the study presents only a single generalized conceptual model based on the results.
The reader might reasonably expect the study to develop conceptual models of wetland–stream connectivity for each of the monitored environments, highlighting their similarities and differences. Instead, the proposed conceptual model is highly generalized and could potentially have been developed from existing knowledge of páramo and puna systems as a hypothesis to be tested through field observations or modeling.
The authors state that one of their main goals is to obtain a process-based understanding that can improve the representation of surface and subsurface processes in hydrological models. However, the conceptual model presented is too generalized to provide a useful representation of a particular high-Andean system with specific climatic, soil, geological, and wetland-development characteristics.
Consequently, the manuscript does not achieve its stated objective. A more useful outcome would be to identify the specific mechanisms controlling wetland–stream connectivity under different environmental conditions and to explicitly distinguish which processes appear to be common across sites from those that are site-specific.
Citation: https://doi.org/10.5194/egusphere-2026-4439-RC2
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 161 | 101 | 28 | 290 | 28 | 24 | 20 |
- HTML: 161
- PDF: 101
- XML: 28
- Total: 290
- Supplement: 28
- BibTeX: 24
- EndNote: 20
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Review of ‘Stream connectivity and governing processes in high-mountain Wetlands‘ by Ross et al.,
Ross et al., present a new study focused on hydrological connectivity, travel times and pathways in 5 South American wetland-stream systems. They use fluorescent (fluorescein) dye injections into the wetland subsurface and passive fluorescein sampling in the connected streams to qualitatively show connection and quantitatively calculate travel times in the wetland systems. This tracer data is interpreted together with groundwater levels and precipitation data. The paper is well written, even if unconventionally structured, on an interesting and important topic, as wetlands are important biomes that are generally underrepresented in hydrological literature. Moreover, there is certainly a bias in understanding wetland hydrology towards the temperate regions of the Northern Hemisphere, and thus studies from the high mountains of the Andes are very welcome. Essentially, they find multiple peaks in their tracer data from the stream, as well as a rapid first peak, which is interpreted as fast overland flow related to fill-and-spill of surface features combined with longer flow paths through the groundwater.
Despite being supportive of the topic and the type of work the authors are conducting in the Andes at high altitudes, I have serious concerns about both the amount of data (very small), as well as the quality of the data used to interpret hydrological processes in the wetlands.
Firstly, to the amount of data: All interpretations about connectivity, transit time and flow pathways are done based solely on the fluorescein injections and the concentrations measured in the stream outlet. There are no basic chemical (major ions), DOC or isotopic (stable water isotopes) or even electrical conductivity data to support the fluorescein tracer data. Also while the authors have measured discharge, this data is not presented in the paper and not used in any way. The lack of any supporting data makes the paper highly speculative, especially when it comes to flow partitioning and transit times. These basic parameters are needed to give any degree of certainty about the processes that are then incorporated into the conceptual model. This is especially relevant for the deep groundwater flow paths, which are impossible to identify in tracer breakthrough curves. The fact that the authors state that this work should provide the basis for methodology that can then be rolled out to networks of catchments (line 95) means that the interpretation needs to be very solid. I also feel that this is why most of the discussion reads like a review paper, with observations from other locations being described in detail because the data doesn’t exist at the site to make any concrete interpretations about what is actually happening at their field sites. It is hard to imagine that the authors went to all the effort of instrumenting and accessing these remote sites and did not take any samples. Therefore, I would encourage the authors to measure any such samples and provide the data to the reader.
Data quality and interpretation: The core of the paper is the travel times: i.e. how long does it take the fluorescein to travel through the wetlands (transit time) and which path does it take between the injection point and the stream. However, this assumes implicitly that there is no interaction between the fluorescein and the organic matrix of the wetlands (mostly peat if I understood correctly) and behaves conservatively in the subsurface over time scales of months. A quick search shows fluorescein adsorbs to organic matter, especially at low pH which is typical for wetlands (e.g. Bork et al., 2020 HESS), which means that the transit times found may simply reflect the interaction between the dye and the organic substrate during transport and differences are related to organic matter content and its composition. Fluorescein was developed for open channels, where residence times are on the orders of hours, and there is limited contact to organic matter. Even in mineral aquifers retardation coefficients are used to correct fluorescein data [1]. I worry here that months in contact with the organic substrate at likely low pH will have substantially retarded the travel times of the dye, leading to incorrect transit times.
Data quality: The authors have exposed activated carbon passive samplers in the streams to sample the fluorescein exciting the wetlands. The samplers have been exposed for ~2 weeks. However, there is no data, evidence or reference as to how quantitively these passive samplers capture the fluorescein either in the lab or in the environment. Passive samplers are seldom perfect, usually with suboptimal recovery rates, especially at high stream flow velocities or if sorption sites are taken by other substances such as DOC. Efficiency also may change with time as sorption sites are occupied, meaning there is no linear relationship between sampling time and concentration in the sampler. Potentially the trends seen are simply a function of sampler efficiency rather than actual changes in concentrations in the stream. Until it is clarified 1. What is the efficiency of these samplers, and 2. How this efficiency is influenced by environmental variables such as stream flow and potentially DOC (which also sorbs to activated carbon) the data are impossible to interpret. I was also thinking that if activated carbon can capture fluorescein so well, why shouldn’t it sorb to the organic matter the wetland as well?
Data quality: dye was still found in the wells months after the experiment started. With a ksat value of ~10-4 to 10-6 m/s and moderate gradients, these wells should have long been flush (assuming diameter of about 5 cm). This makes me wonder if the well screens were connected to the wetland groundwater or potentially clogged during installation.
Analysis: I think it would be very helpful to look at mass fluxes and construct something like a residence times distribution (RTD). By multiplying concentrations (albeit averaged) with the flow, you have a mass flux (g/s). Dividing by total mass added (g) and plotting over time you have a RTD of tracer in the wetland. The useful thing here is that it is independent of flow conditions and things such as dilution.
Interpretation: The core of the interpretation is that the fill and spill mechanism is responsible for the rapid peaks and the subsequent peaks due to flow through the groundwater or other deeper flow paths. Assuming we can trust the data, this behaviour could just as easily be explained by the ‘transmissivity feedback mechanism’ [2,3], whereby the hydraulic conductivity decreases exponentially with depth in the peatland. As the water table rises the runoff increases in accordance with exponentially increasing ksat. Variable water table can then lead to multiple peaks. This is at least something to consider given the lack of complementary data to assist in the interpretation of variable flow paths contributing to stream flow.
Other points:
Figures: representing the data as a line graph gives the impression of a continuous dataset. As the authors use passive samplers to integrate fluorescein concentrations over a set time interval, a more accurate way to plot the data as horizontal bars representing the exposure period, without the continuous lines.
There are many in-text references, often in long lines. This often makes it hard to follow the actual text of the paper. I would recommend reducing to maybe three references that really capture what the author wants to say.
How it is possible to have a porosity of 0.88 and a hydraulic conductivity of 10-7 m/s-10^-4 m/s? There is only 12% mass!
It could be made a little clearer that µg/L/d means the concentration divided by the exposure time, and this assumes that the concentration on the activated carbon increases linearly with time (which I am very unsure about without more data).
Much of what you have suggested should be implemented in models has been done to quite a large extent by e.g. Frei et al., [4,5] using hydrogeophere.
Refs
[1] Sabatini, 2000, Sorption and intraparticle diffusion of fluorescent dyes with consolidated aquifer media, Ground Water.
[2] Kendall et al., 1999, A hydrometric and geochemical approach to test the transmissivity feedback hypothesis during snowmelt, USGS.
[3] McGuire et al., 2024, JAMES BUTTLE REVIEW: Interflow, subsurface stormflow and throughflow: A synthesis of field work and modelling. Hydrol. Pross.
[4] Frei et al., 2010 Effects of micro-topography on surface–subsurface exchange and runoff generation in a virtual riparian wetland — A modeling study, Advances in Water Resources.
[5] Frai et al., 2012. Surface micro-topography causes hot spots of biogeochemical activity in wetland systems - A virtual modeling experiment