the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Taking the pulse of nature – How robotics and sensors assist in lake and reservoir management
Abstract. Ecosystems, like almost any environmental entity, are often highly sensitive to the presence of humans when measuring field characteristics. Robotic solutions deserve attention to avoid or greatly reduce related bias. Constant availability of robotic solutions, independent of the time of day and most weather conditions, is an additional advantage.
Here, we present an autonomous, Modular Aquatic Robotic Platform (MARP-FG) designed to collect relevant environmental information from surface waters. We define the demands, describe the encountered obstacles and how to overcome them. MARP-FG implements autonomous navigation and data collection capability across various floating-body configurations and sensor setups. Depending on the weight of the measurement system (payload), catamaran floaters with a length ranging from 1.2 meters to 2.5 meters are used. We realized and evaluated three different payloads based on the MARP-FG concept: i) Hydrographic profiling with a multi-parameter probe, ii) Sonar-based 3D mapping of complex basins, and iii) Dynamic closed chamber-based greenhouse gas exchange determination with on-board CO2 quantification (IR spectrometry) and gas sampling (Exetainers®) for subsequent gas-chromatographic analysis.
This work focuses on option iii) as a practical example to describe our design process and operational modes, thus minimizing faults and errors, especially in harsh environments. Full operation was possible to wave heights of ±40 cm and wind speeds to 7 m sec-1. Positioning accuracy during measurement cycles was on average better than ±2 m in xy directions. The platform has demonstrated its capabilities in field campaigns on lakes in the Amazon basin (Brazil) and on waterbodies in temperate climate regions of Europe. Largely improved and reproducible positioning on a waterbody, full functionality also under adverse weather conditions and during nighttime significantly enhanced high-quality data acquisition and opens new applications.
- Preprint
(1215 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on egusphere-2024-2261', Anonymous Referee #1, 29 Sep 2024
This article explores the use of an autonomous robotic platform (MARP-FG) for the monitoring of lakes and reservoirs, which is a field of interest. With global climate change and environmental degradation, accurate monitoring of the health of water bodies and ecosystem services has become increasingly important. The use of robotic and sensor technology can reduce human bias and improve the frequency and accuracy of data collection, with data and related details uploaded to GitHub. After considering the following questions, I recommend it for publication.1. Has the article fully considered all factors that may affect the measurement results, such as water temperature and weather changes? In data analysis, have environmental factors such as water temperature and weather changes been controlled or corrected to ensure the accuracy of the measurement results?2. Has the platform undergone long-term deployment testing to assess its stability under environmental changes over an extended period (longer than one year)?3. Has the platform been tested under extreme environmental conditions (such as storms, floods, extreme temperatures, etc.) to evaluate its stability?4. Has testing been conducted in more situations, such as high-altitude or high-latitude regions?5. Any measurement may have an impact on the object being measured. Although the article emphasizes that the equipment used has minimized the impact, is there a quantitative method to test the impact of the equipment on the environment measured? If possible, add a control group experiment to compare the effects of traditional monitoring methods with MARP-FG.6. Has the applicability of MARP-FG in different types of water bodies (such as freshwater lakes, saltwater lakes, etc.) been considered? Is the platform's performance in saltwater bodies comparable to that in freshwater bodies? Will the salinity of saltwater affect the accuracy and lifespan of the sensors? Are there any sensors or protective measures specially designed for saltwater environments?Citation: https://doi.org/
10.5194/egusphere-2024-2261-RC1 -
CC1: 'Reply on RC1', Sebastian Zug, 01 Oct 2024
Many thanks to the referee for the feedback and questions about the paper. Subject to the reviewers' findings, we reply as follows.
1. Has the article fully considered all factors that may affect the measurement results, such as water temperature and weather changes? In data analysis, have environmental factors such as water temperature and weather changes been controlled or corrected to ensure the accuracy of the measurement results?
In addition to the sensors of the navigation unit (GNSS, IMU, compass), the robot itself comprises the sensor systems listed in Table 1. Measurement parameters such as water temperature, humidity, wind direction and wind speed are continuously recorded with the respective accuracies given in Table 1. An influence of the parameters could not be shown in the experiments. No deviations beyond the measurement errors documented in the data sheets occurred during the testing of the sensors under laboratory conditions.
2. Has the platform undergone long-term deployment testing to assess its stability under environmental changes over an extended period (longer than one year)?
The concrete platform itself was developed as of 2019 with intensive consecutive testing of the platform (still without robotic capabilities) on water bodies in Germany. We then successfully applied it in 5 campaigns (2021–2023) in the Amazon region. Improvements were made continuously within and following each campaign, so that a long-term test of the current setup has not yet been carried out. In parallel, a twin platform was used on Central European water bodies; results have been published already (https://www.tandfonline.com/doi/ full/10.1080/20442041.2024.2388339?src= <https://www.tandfonline.com/doi/ full/10.1080/20442041.2024.2388339?src=>)
3. Has the platform been tested under extreme environmental conditions (such as storms, floods, extreme temperatures, etc.) to evaluate its stability?
Yes, extreme weather conditions – maximum humidity (>90%), high temperatures (>40 degrees Celsius) and intense rainfall (>50 mm per hour), and complete darkness – were encountered during the measurement campaigns in the Amazon. To meet these challenges, the team implemented various improvements. For example, sun protection covers for electronic boxes were tested, as were cooling systems for the integrated computers. The team integrated a flap system to calm the water between the floats in the measuring area. This allows the waves in the gas bell to be reduced. The article manuscript points this out.
4. Has testing been conducted in more situations, such as high-altitude or high-latitude regions?
No, we have not tested the platform in high-elevation (e.g., high mountain lakes) or high-latitude (e.g., polar environments) yet. However, we are confident that neither boundary conditions pose serious challenges to the current setup. The platform was successfully tested in Central European winter with below freezing temperatures.
5. Any measurement may have an impact on the object being measured. Although the article emphasizes that the equipment used has minimized the impact, is there a quantitative method to test the impact of the equipment on the environment measured? If possible, add a control group experiment to compare the effects of traditional monitoring methods with MARP-FG.
Since this is not feasible for the gas exchange setup, we restricted related testing to hydrographical profile acquisition. Here, we made parallel determination from the platform and manually from a boat nearby – yet not on Amazon lakes. These tests showed higher accuracy and precision for the platform- derived data.
6. Has the applicability of MARP-FG in different types of water bodies (such as freshwater lakes, saltwater lakes, etc.) been considered? Is the platform's performance in saltwater bodies comparable to that in freshwater bodies? Will the salinity of saltwater affect the accuracy and lifespan of the sensors? Are there any sensors or protective measures specially designed for saltwater environments?
In saltwater scenarios, components are always subject to greater wear. The larger version of the MARP-FG system was used in Croatia in the Mediterranean in 2023 to support the tracking of divers. After the 10-day campaign, the entire platform had to be dismantled and thoroughly cleaned. No damage was detected at any sensor.Citation: https://doi.org/10.5194/egusphere-2024-2261-CC1
-
CC1: 'Reply on RC1', Sebastian Zug, 01 Oct 2024
-
RC2: 'Comment on egusphere-2024-2261', Anonymous Referee #2, 19 Nov 2024
The paper discusses the development and implementation of a robotic platform, MARP-FG, designed for environmental monitoring, specifically in lakes and reservoirs. The study highlights MARP-FG’s advantages, such as its autonomy, robustness in harsh conditions, and the ability to perform measurements with minimal human intervention, which reduces bias and enhances safety.
Suggestions for improvement:
- Clarify the comparison between MARP-FG and other robotic platforms in environmental monitoring to emphasize unique contributions.
- Provide more quantitative results on data quality improvements achieved with MARP-FG (e.g., stability metrics in greenhouse gas measurements compared to manual methods).
- Discuss any potential environmental impacts of the platform itself, such as energy use and interactions with local wildlife. How does MARP-FG minimize its impact during deployment in sensitive ecosystems?
- Expand details on MARP-FG's modular adaptability in different missions, as this is a core strength.
- Given the energy-intensive tasks (e.g., 3D sonar mapping), detailing power efficiency strategies, battery hot-swapping procedures, or even potential renewable energy options (such as modular solar panels) could enhance the platform’s operational range and environmental sustainability.
- You could provide a more discussion on MARP-FG's limitations in handling extreme environments. For instance, specifying the upper limits for humidity, temperature, or water turbulence where MARP-FG can still function optimally could help clarify its resilience.
- To improve data consistency across different mission types, establishing standardized data formats and describing these in the paper would aid researchers in efficiently managing and analyzing data from different payloads.
- Some of the hyperlinks in the paper are not functional.
- On page 9, line 367, there's an incorrect reference to a figure. The text is explaining Figure 6, but it mistakenly says Figure 5.
Questions :
- Why was there no detailed statistical analysis or interpretation of the greenhouse gas flux data presented, especially given the claimed advantages of continuous data collection by the robotic platform?
- Given that the paper focuses on the reliability and accuracy of robotic deployment, what additional analyses could be conducted to demonstrate the environmental implications of the collected measurements, such as correlations between gas fluxes and environmental conditions?
- What specific limitations or uncertainties exist in the gathered greenhouse gas data, and how could future studies leverage data analysis techniques to provide clearer environmental or biogeochemical conclusions?
Additional questions:
- How do you plan to address limitations in MARP-FG’s error communication and mitigation strategies for nighttime and long-distance missions?
- How does MARP-FG’s design account for climate-induced changes, such as fluctuating water levels or extreme weather patterns? Have stress tests been conducted to ensure performance in these variable conditions?
- How is long-term data storage handled, especially for multi-season campaigns in remote areas? What strategies are in place to ensure data continuity and accessibility over extended periods?
- Could multiple MARP-FG platforms work together in coordinated tasks to cover larger areas more efficiently? If so, what communication protocols would be necessary to facilitate this collaboration?
Citation: https://doi.org/10.5194/egusphere-2024-2261-RC2 -
AC1: 'Comment on egusphere-2024-2261', Jörg Matschullat, 21 Nov 2024
Attached, please find our responses to the referees comments. We made a few amendments to the original manuscript to help clarify some points and to correct the obviously wrong link (Text reference to figure 6 instead of figure 5). Kindly advise us if we can upload that corrected version of if we shall make the necessry correction in the pre-print (proof) document?
Sincerely
Jörg Matschullat for all authors
-
AC2: 'Comment on egusphere-2024-2261', Jörg Matschullat, 05 Dec 2024
Dear Editors and dear referees,
We appriate your efforts to process our manuscriupt and thank you for the comments. Mostly your questions motivated us to revise the manuscruipt in respect to language, since we got the impression that some misunderstandings occured. W eno whope toin find the proper pathway for the revised manuscript...
Uploaded here is a short document with replies to the referees, hoping that all of their questions could be satisfactorily answered.
In the name of Sebastian Zug and myself, as well as all co-authors, we wish you a beautiful, peaceful holiday season and a good start in a healthy, and properous Year 2025
Jörg Matschullat
Viewed
HTML | XML | Total | BibTeX | EndNote | |
---|---|---|---|---|---|
148 | 51 | 185 | 384 | 8 | 9 |
- HTML: 148
- PDF: 51
- XML: 185
- Total: 384
- BibTeX: 8
- EndNote: 9
Viewed (geographical distribution)
Country | # | Views | % |
---|
Total: | 0 |
HTML: | 0 |
PDF: | 0 |
XML: | 0 |
- 1