IoT Analytics and Its Application in Marine Aquaculture

By Assoc. Prof Dr. Nurul Hashimah Ahamed Hassain Malim (School of Computer Sciences)

November 2021 VOICES OF USM
main image
An effect of eutrophication.
Advertisement

USING IoT ANALYTICS, the Schools of Computer Sciences and Chemical Engineering at Universiti Sains Malaysia are working together with the fish farming community of Sungai Udang to minimise the risk of fish kill1 from eutrophication, a condition where water columns are rich in nutrients (particularly phosphorus and nitrogen) from regular fish feeding and the discharge of fertilisers or sewage into the aquatic system.

Eutrophication causes oxygen depletion in coastal environments and stimulates Harmful Algal Bloom (HAB). Where the HAB is non-toxic, the fisheries resources will only suffer some damage, but if the reverse happens, the entire aquatic system is threatened.

Water turbidity also increases with algae growth; it blocks light from passing through and narrows the light spectrum below water surface. These altered light conditions have been shown to affect the reproductive behaviour of fishes and indirectly foiling fisheries production.

Readings of parameters such as water and ambient temperature, turbidity, solar radiation, total suspended solids, pH levels and salinity all previously required manual sampling and in situ observation of bloom occurrences. But these can now be facilitated through IoT analytics.

In its simplest function, IoT connects devices that are wired to the internet.1 But an elaborate, more interconnected web for data generation and sharing is formed when wireless sensors, software, actuators and electronic devices are linked together for remote control and monitoring.

The original form of the water station powered by solar energy.

Predicting Conditions

When combined with analytics, it sifts through vast amounts of heterogeneous data to offer insights, break down statistics to study data trends and patterns, predict the probability of events or conditions to alert the user, and even to recommend actionable solutions.

But IoT analytics does more than just read and process data; it reacts to them by way of response mechanisms. These post-analytics actions are done either manually, with automation (robotics process automation) or include interfacing to an external system that controls the equipment used for response action upon event detection.

Actuators facilitate this process, converting electrical signals into physical movements whenever there is a need to trigger other devices or machines into operation. Take the example of a fire sprinkler system, when an abnormal level of heat is detected in a building, the heat sensor communicates this to the monitoring system (or the control / command centre), which then prompts the sprinklers to be turned on. The actuator, either the pneumatic or solenoid variant, reacts on activation by pushing or pulling the valve of the fire sprinkler in which it is fitted.

To monitor oxygen levels and algae growth of the cagefarm at Sungai Udang, USM researchers have created an IoT-operated, solar-powered water station affixed with six different sensors, for pH, turbidity, dissolved oxygen, temperature, electrical conductivity and total dissolved solids. These are mounted to a microcontroller (Arduino Bluno) that connects to the internet.

Per minute data is tracked and stored in cloud storage, from which it is extracted by the analytics platform Water Quality Monitoring System (WQMS) to monitor sudden drops of oxygen level. The system is also equipped with predictive analytics to determine when oxygen levels start to dip in response to cycles of algae growth. This prepares farmers for manual response mechanisms to be taken.2

USM researchers during a visit to the fish farm at Sungai Udang.

In the near future, the WQMS will be linked to the fish farm aeration system to ensure a swift response if depletion of oxygen levels is detected in the water, and to avoid accidental fish kills. But linking the two systems is conditional to the availability of the aeration system, and its flexibility or compatibility for interfacing with the WQMS.

For this project, low-cost sensors powered by solar energy are used as an alternative to costlier commercial ones, with a raw cost of not more than RM2,000. The retail price for sensors ranges around RM20,000 and comes only with standard statistics-based software for monitoring. Even with the addition of more features, this raw cost would still be affordable for fish farmers.

Presently, the researchers are still experimenting with the set-up, and finding solutions to problems encountered in every installation of the water station. The exact lifespan of the system is yet undetermined.

To know more about the project, go to dl.acm.org/ doi/10.1145/3416921.3416928 or watch the ASEAN IVO Forum 2020 at youtu.be/RFbM3QwcXac

Acknowledgment

We would like to thank the fish farming community in Sungai Udang for the opportunity given. This project is funded by the Ministry of Higher Education Malaysia, under the Trans-Disciplinary Research Grant Scheme (TRGS) (203/PKOMP/67612004). Other researchers involved in the TRGS are Prof Dr. Abdul Latif Ahmad, Prof Dr. Ooi Boon Seng, Assoc. Prof Dr. Derek Chan Jiunn Chieh, Prof Dr. Rosni Abdullah, Mohd Azam Osman, Nur Aqilah Paskhal Rostam (Postgraduate Student), and Abdul Aziz Abdul Hadi (Undergraduate Student).

References
  • K. Patel, S. M. Patel, and P. G. Scholar (2016) Internet of Things-IOT: Definition, Characteristics, Architecture, Enabling Technologies, Application & Future Challenges. International Journal of Engineering Science and Computing 6(5): 1-10.
  • S. Sedkaoui (2018) Data Analytics Process: There’s Great Work Behind the Scenes in Data Analytics and Big Data, WILEY.
  • https://www.lifewire.com/fitbit-sleepstages- 4140333
  • https://www.twi-global.com/technicalknowledge/ faqs/vibration-analysis
  • https://realpars.com/actuator/
  • Chislock, M. F., Doster, E., Zitomer, R. A. & Wilson, A. E. (2013) Eutrophication: Causes, Consequences, and Controls in Aquatic Ecosystems. Nature Education Knowledge 4(4):10
  • Anderson, D.M. (2009) Approaches to monitoring, control and management of harmful algal blooms (HABs). Ocean Coast Manag52(7): 342.
  • https://www.conserve-energy-future.com/ causes-effects-and-solutions-to-eutrophication. php
  • Jarvenpaa, M. & Lindstorm, K. (2004) Water turbidity by algal blooms causes mating system breakdown in a shallow-water fish, the sand goby Pomatoschistus minutus. Proceeding Bological Science 271 (1555):2361.
  • Harmful algal blooms and climate change: Learning from the past and present to forecast the future. Harmful Algae 49: 68–93.
  • Huang, J.D., Zheng, H.R. (J.) & Wang, H.Y. (2017) Current Trend of Metagenomic Data Analytics for Cyanobacteria Blooms. Journal of Geoscience and Environment Protection 5:198- 213.
  • Whitehead, P.G., Howard, A. & Arulmani, C. (1997) Modelling algal growth and transport in rivers: a comparison of time series analysis, dynamic mass balance and neural network techniques. Hydrobiologia 349: 39
PM
Assoc. Prof Dr. Nurul Hashimah Ahamed Hassain Malim (School of Computer Sciences)

specialises in data analytics. She is keen on trying out different application areas to benefit from her ever-growing interest in data analytics.