Groundwater Level Assessment Using Smartphones (2) (2)

1. ASSESSING GROUNDWATER LEVELS in WELLS USING SMARTPHONE SENSORS

Yonatan Habtom Mhreteab &
Freddy Gitare
Supervisor : Dr.-Ing. Fritz Kalwa
Institute for Groundwater Management - TU Dresden
Study Project Final Presentation
Jul 17, 2025
1

2. INDEX

1
Objectives
4
Discussion
2
Results - GGA
5
References
3
Results – Sonar Scripts
2

3. OBJECTIVES

1. OBJECTIVES
Overall Aim: To advance smartphone-based
groundwater level measurement
• Quantify accuracy & precision of GGA across diverse
phones, and well conditions.
• Design, develop & validate a smartphone sonar script
(Python prototype) for depth measurement.
3

4.

2. Result GGA
Experiment Setup
Simulated wells using PVC pipes of depths :
0.5m, 2.05m, 6.05m, 12.05m, 20.13m.
18 different phones, 21 measurements per phone:
Total of 378 measurements recorded.
Uses Acoustic resonance to estimate depth
4

5.

2.1. Results - GGA
Fig 1. Mean Error in Depth Estimation by Phone Type
5

6.

2.2 Results - GGA
Purity in GGA is a metric used to measure the confidence in the estimated depth
Fig 2. Purity of Measurement by Depth and Tune
6

7.

3. Result SONAR script
Sonar Script development steps
Depth range categorization & choice of appropriate signal pulse
Auto Correlation Analysis
Cross Correlation Analysis
Echo detection and time of flight estimation
Depth estimation
Estimation accuracy analysis
7

8. 3.1 Results – Categorization and Pulse Design

Pulses was designed to maximize echo detection.
0.5 m – 2.0 m Depth
2.0 m – 5.0 m Depth
5.0 m – 10.0 m Depth
10.0 m – 30.0 m Depth
Fig 3. Sonar pulses Suite for Well Depth Analysis
8

9. 3.2 Results – Autocorrelation Method

Estimation of test script on 2.05 m depth pipe – Autocorrelation Method
True Depth => 2.05 m
Estimated Depth => 2.0 m
Fig 5. Raw Signal and Autocorrelation method of estimation for 2.05 m pipe
9

10. 3.3 Results – Cross Correlation Method

Estimation of test script on 6.05 m depth pipe – Cross correlation Method
Fig. Design Pulse
True Depth => 6.05 m
Estimated Depth => 6.0 m
Fig 5. Raw Signal and Cross correlation method of estimation for 6.05 m pipe
10

11. 3.4 Remaining Steps – SONAR script

Fine tuning of the script to estimate greater depths
Developing a Confidence Metric like the Purity metric used in GGA
using a consensus algorithm combining the auto correlation and
cross correlation methods
Field Testing
11

12. 4. Discussion

Shorter pipes had lower MAE (better accuracy) but more inconsistent purity,
likely due to shallow signal noise
Longer pipes had higher MAE but more stable, higher purity, showing better
noise filtering at greater depths.
The modest positive correlation between depth and purity suggests the app
performs better in deeper environments.
Sonar Script shows good estimation for depths up to 6m with acceptable error so
far, but it is prone to environmental noise in its performance.
30 m pipes are found to be challenging for both the GGA and Sonar app.
12

13.

Start: Project Setup and
Preparation
PROGRESS
Core Research Activity
Step 1 - Evaluate Groundwater
Global App
Step 2 - Develop and Validate
Python Sonar Script
1a. Field Testing
2b. Script Development
1b. Lab Testing - Dresden:
controlled setup
2c. Lab & Field Validation
1c. Analyze GGA Data Accuracy, Uncertainty, Factors
2d. Analyze the algorithm for
accuracy, uncertainty
2a. Data Collection
Completed Tasks
Yet be done
Step 3 - Analysis and Synthesis
Step 4 - Conclusions and
Reporting
Formulate Recommendations
and Feedback for GGA
Prepare Final Report and
Presentations
End of Project Deliverables
Partially completed

14.

Field Measurement
in Pirna
14

15.

6
References
Benjamin M.Faber (2017). Acoustical Measurements with Smartphones: Possibilities and Limitations: A
smartphone-based sound level meter or analyzer may or may not replace your expensive, precision
instruments.
Brkic, M., Obradovic, D., Dudarin, Z., Károly, B., & Živanov, M. B. (2013). Measurement and Monitoring
System for Level of Groundwater. Key Engineering Materials, 543, 243–246.
https://doi.org/10.4028/www.scientific.net/KEM.543.243
Elias, M., & Maas, H.-G. (2022). Measuring Water Levels by Handheld Smartphones: A contribution to exploit
crowdsourcing in the spatio-temporal densification of water gauging networks. The International Hydrographic
Review, 27, 9–22. Retrieved from https://journals.lib.unb.ca/index.php/ihr/article/view/33130
Groundwater Global Foundation (2025, February 7). Retrieved from https://www.groundwater.global/
Muste, M., Fujita, I., & Hauet, A. (2008). Large‐scale particle image velocimetry for measurements in riverine
environments. Water Resources Research, 44(4). https://doi.org/10.1029/2008WR006950
15

16.

THANK YOU
Q&A
16