the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
River intermittency: mapping and upscaling of water occurrence using unmanned aerial vehicle, Random Forest and remote sensing landscape attributes
Abstract. Although intermittent rivers exist naturally, global changes have a direct influence on streamflow permanence. Measurements and modelling in temporary rivers are still scarce and yet, essential for prediction and understanding of scarcity scenarios. Thus, this work aims to map and model the spatio-temporal dynamics of an intermittent river. The study area is the Umbuzeiro River in the Brazilian Semiarid (∼100 km), whose spatially coherent streamflow occurs exclusively in the wettest months during the rainy season. We conducted twelve UAV surveys between March and November 2022 in selected river reaches. With the imagery from UAV surveys, we classified river reaches into "Wet", "Transition", "Dry" or "Not Determined" with visual inspection of 1.0 m reaches. In order to explain the observed patterns, we analysed 40 candidate predictors based on static and dynamic landscape attributes. Among these, altitude, drainage area, distance from dams, and dynamic predictors proved to be most informative in Random Forest models. We selected three Random Forest models based on the different dynamic predictors. The models differ in the source and type of dynamic predictor used to capture the temporal dynamics: (a) series of Sentinel MNDWI; (b) series of Planetscope NDVI; and (c) accumulated precipitation (30 days). All model variants successfully mimicked river intermittency with an accuracy of around 80 % for both test and training. Models (a) and (b) captured the temporal dynamics in model extrapolation to the whole river. When analysing the spatial distribution of intermittency, models (a) and (c) better identified areas more prone to "Wet" or "Transition" classes. This way, model (a) was identified as the most successful in simulating intermittency both temporally and spatially. The use of Sentinel MNDWI in model (a) aggregates enough spatial information, so the model can better simulate water occurrence classes. The findings presented here emphasize the possibility of using this index even in narrow temporary rivers. The results provide insight into the hydrological diversity of semi-arid rivers and are, therefore, important to understand their role in water availability.
- Preprint
(38962 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on egusphere-2025-884', Anonymous Referee #1, 30 Jul 2025
General Information
The study investigates spatial and temporal dynamics of river intermittency in the Umbuzeiro River (Brazilian semiarid region) using a combination of UAV surveys, remote sensing, and Random Forest (RF) modeling. The authors assess water presence classes: “Wet”, “Transition”, “Dry”, and “Not Determined”, across different times and river segments. They evaluate three RF model variants based on different dynamic predictors: Sentinel MNDWI, Planetscope NDVI, and 30-day accumulated precipitation.
This is a well-structured and technically rigorous study that addresses a critical gap in intermittent river monitoring in semi-arid environments. It provides an important proof of concept for combining UAV-based mapping and machine learning for ecohydrological research. The results are promising and provide a methodological blueprint for future upscaling efforts, though there is room for refinement in automation, validation, and generalization.
Despite the interest in the topic, I believe some changes in the paper are needed before it can be considered ready for publication, listed in the specific comments.
Specific Comments
- LL121-123:"The "Not Determined" class included those reaches where it was not possible to discern”. It may be useful to evaluate the integration of soil moisture indices or evapotranspiration estimates as additional dynamic predictors.
- L143: “These indices are are summarized in Table 1”. The two verbs reported need to be corrected.
- L189-192: “Damming structures are mapped all along the Umbuzeiro River by using Planetscope images so as to visually locate each dam.” The river path and dam structures are mapped manually using high-resolution imagery and field data. In particular, the geolocation of the dams was also performed manually, which limits scalability and reproducibility for larger or other basins. Could automated surveying using satellite data be considered?
- L257: “4.1 Observed water intermittency: UAV imagery”. From the text, it appears that the UAV-based classification is visual and not verified with in-situ hydrological measurements or ground truth sampling. The lack of objective thresholds could introduce bias in the classification of “Transition” and “Wet” classes. Probably longer UAV survey campaigns, covering more years, could better capture interannual variability.
- LL326-328: “Temporal variations along the year can also be observed in comparison to monthly precipitation.” What is the period used to calibrate the model to identify the seasonality of events? The model using 30-day precipitation (Model c) underperforms in capturing seasonal transitions. This approach limits the model’s ability to generalize across varying wet/dry years.
- LL351-353: “Model (a) is even more specific in this respect and indicates mainly areas in the lowest part of the basin. The identification of areas prone to wetter conditions is very important even in the smallest of scales because they can be key areas for river ecology, for instance”. Although model (a) achieved the best results, the use of Sentinel MNDWI data may not be generalizable to narrower or canopy-covered watercourses, especially in forested catchments.
- L378: “for spectral indexes based on UAV data”. It's better to use indeces. "Indexes" is commonly used to refer to alphabetical lists in books, for example. In contrast, "indeces" is used in more technical, scientific, and mathematical contexts.
- L394: “Conclusion”. Probably the conclusion could be more detailed and not a simple summary of the study done; they could report some details shown in the graphs and tables.
Citation: https://doi.org/10.5194/egusphere-2025-884-RC1 -
RC2: 'Comment on egusphere-2025-884', Anonymous Referee #2, 27 Aug 2025
This study addresses a critical gap in intermittent river monitoring in the Umbuzeiro River, Brazil, which I consider significant with implications for other arid areas and for anticipating future climate-change impacts. The integration of UAV surveys, Random Forest classification, and remote sensing–based landscape attributes is a valuable approach. However, the manuscript in its current form suffers from presentation and clarity issues that obscure the scientific contributions. Substantial revision is needed to improve communication, sharpen the articulation of the scientific questions more process-oriented interpretation of results in the context of Earth surface processes.
Major Comments
- Scientific framing
- While the methodological framework is carefully outlined, the manuscript must articulate the scientific question more strongly. The introduction hints at this, but the discussion largely focuses on model outcomes without interpretation in terms of physical or hydrological processes. For example: “distance from the last dam is the most important predictor” does not require a complex Random Forest model to identify. Similarly, Section 4.2 notes that elevation and drainage area are important predictors, but these are already well established. This journal is not a primarily GIS/remote sensing journal and so I encourage the authors to discuss What is the added value of this modeling exercise? Does it allow optimization of resolution? What unexpected findings emerged? What are the implications for other arid environments or climate-driven changes in intermittency?
- Presentation quality
- The manuscript follows the standard structure, but the communication of ideas and concepts requires improvement. At present, several sections read like figure captions rather than explanatory text (e.g., line 219). Statements should be fine-tuned for accuracy and consistency (e.g., lines 109–111).
- Section 3.1 Modelling workflow requires rewriting: the repeated use of the generic term “data” is too vague. Please specify clearly whether you are referring to UAV-derived classifications or predictor variables.
- Figures require improvement. For example, continuous variables such as monthly precipitation should be plotted as interpolated lines with points not bar plots. Figure captions must be complete and self-contained.
- Methods
- Although all equations are listed, the conceptual meaning of the metrics (e.g., balanced accuracy) is not unpacked, remember your audience. Please provide intuitive descriptions and, if possible, a conceptual diagram.
- Please clarify why Table 2 lists 25 predictors, but Fig. 8 refers to 40.
- Results and interpretation
- Some results are not fully explained. Why does Model (c) predict >75% dry in March, peak wet season, but not in November, second wet season? Why is Model (a), with 5-day frequency predictors, described as the most successful in simulating intermittency when it has the coarsest temporal resolution?
- Please interpret results in physical terms, not only in terms of model metrics.
- References and context
- The references are strongly weighted toward recent Brazilian studies. Please include some of the classic papers on ephemeral and intermittent streams.
- The paper should also acknowledge interannual variability in flow connectivity, which is critical for interpreting results beyond a single year.
Minor Comments
Lines 109–111: Be consistent in describing rainy season timing. April–July are not “the first months of the year.”
Line 134–135: Do you need to direct readers to the Sentinel user guide?
Lines 135 & 161: Clarify the role of Google Earth Engine (GEE) for Sentinel and MapBiomas.
Line 140: Specify what was done in R.
Line 165: Where in Figure 1 is this shown?
Line 172: Please provide information on data access (can it be downloaded and where?).
Line 203: Suggested rewording: “containing the type of variable…”
Line 207: Cite Table 2 after “candidate predictors.”
Line 219: Reword to “The specific predictors employed in each model are summarized in Table 2.”
Line 360: Section 4.6 Add acknowledgment of interannual variability and how the intermittency might vary
Line 419: Avoid anthropomorphizing landscapes; specify which aspects of river metabolism are relevant.
Figures
- Fig. 1: Add discharge/precipitation panel to establish arid context, I know precipitation data is presented in Figure 9
- Fig. 2: Show/explain how exactly the UAV surveys are used
- Fig. 3: Quantify misfit between flow accumulation (DEM) and visual mapping. Adjust blue line color for visibility.
- Fig. 4: Add discharge/precipitation to support text (lines 109–110).
- Fig. 5: Missing panel label (a). Clarify “1.0 m reaches,” as 1 m seems implausibly small.
- Fig. 7: Clarify why parameters with high Mean Decrease Accuracy were selected.
- Fig. 8: Please clarify why Table 2 lists 25 predictors, but Fig. 8 refers to 40.
- Fig. 9: Explain Model (c) behavior, see Major comments.
- Fig. 10: Clarify how “High” and “Low” were calculated. Can values be normalized and expressed as percentages?
Tables
- Table 2: Organize predictors by model (a–c), consistent with Figs. 7, 9, and 10. Replace “constant” as that is not a frequency
- Table 3: Define balanced accuracy conceptually, I know the equation is given, and present Train/Test values in a clearer layout.
Data & codes
Please include the data and codes to comply with ESurf's FAIR policies
Citation: https://doi.org/10.5194/egusphere-2025-884-RC2
Viewed
HTML | XML | Total | BibTeX | EndNote | |
---|---|---|---|---|---|
366 | 51 | 14 | 431 | 8 | 22 |
- HTML: 366
- PDF: 51
- XML: 14
- Total: 431
- BibTeX: 8
- EndNote: 22
Viewed (geographical distribution)
Country | # | Views | % |
---|
Total: | 0 |
HTML: | 0 |
PDF: | 0 |
XML: | 0 |
- 1