Page 17 - AJWEP-22-4

IoT-based water quality monitoring

Based on Table 3, data on Thursday were collected Table 4. Based on Table 4, on Monday, the highest error
on the 1 day of the river wastewater monitoring tool on data were 1.3%, and the lowest error data were 0.3%.
st
the Buntu River in Kendal Regency. From Thursday to On Tuesday, the highest error data were 1.9%, and the
Friday, it can be seen that the error data are 0.99%. On lowest error data were 0.3%. On Wednesday, it can be
Saturday to Sunday, the highest error data were 1%. The seen that the highest error data were 1.9%, and the lowest
lowest error data were 0.99%, and the Monday showed error data were 0.6%. On Thursday, the highest error
that the highest error data were 1%, whereas the lowest data were 1.6%, and the lowest error data were 0.0%.
error data were 0.97%. The values obtained from the pH On Friday, it can be seen that the highest error data were
sensor with the pH meter measuring tool were similar 1.5%, and the lowest error data were 0.3%. The values
at the same time. The average result of testing standard obtained from the temperature sensor with the digital
error in IoT was 0.99%. The standard error between temperature measuring tool are similar at the same time.
the pH sensor and the pH meter measuring tool, at The standard error between the temperature sensor and
approximately 2% was declared valid. the digital temperature measuring tool, at approximately
The quality of river water is strongly affected by pH, 2%, was declared valid. Based on Table 5, on Thursday,
which influences the solubility of metals, water alkalinity, the highest error data were 1.01%, and the lowest error
and microbial metabolism. Typically, the uptake of data were 0.98%. On Friday, the highest error data were
dissolved carbon dioxide by photosynthetic algae raises 1%, and the lowest error data were 0.98%. On Saturday,
pH levels. Conversely, rivers contain large quantities of it can be seen that the highest error data were 1.01%,
organic matter, including colloidal suspensions, which and the lowest error data were 0.99%. On Sunday, the
often display acidic properties. Moreover, the release data showed that the highest error data were 1%, and
of domestic and industrial wastewater can negatively the lowest error data were 0.98%. On Monday, it can
impact pH levels in the aquatic ecosystem. 51 be seen that the highest error data were 1.01%, and the
lowest error data were 0.99%. The values obtained from
3.2. Temperature the temperature sensor with the digital temperature
Temperature readings also demonstrated strong measuring tool are similar. The average result of the
consistency between the IoT sensor array and digital testing standard error in IoT was 0.99% at the same
thermometers. In Banjardowo river, the error rate time. The standard error between the temperature
varied between 0.3% and 1.9%, while in Buntu river, sensor and the digital temperature measuring tool, at
it remained within a narrow band of 0.98 – 1.01%. approximately 2%, was declared valid.
These results were in line with previously published The highest temperatures here may be related to
evaluations of water temperature sensors in IoT the depth of the water compared to other rivers. The
systems, which typically report accuracy within ±0.5°C addition of waste and increased anthropogenic activities
under field conditions. 49,52 Slight discrepancies may near these sites may also be considered causes of the
arise due to variations in water mixing, shallow depth temperature increase. Human-caused disturbances such
exposure, or direct solar radiation on the sensor housing. as urbanization and waste dumping have significantly
However, the average error of <1.5% remains within changed the temperature of water bodies, which has
acceptable limits for most aquatic ecosystem studies, also impacted flora and fauna. Water temperature, which
as critical biological processes such as DO saturation, plays a vital role in limiting oxygen content, emerged
54
metabolic rates, and nutrient solubility follow broader as a crucial parameter within this sub-catchment,
thermal trends rather than precise thresholds. Given significantly influencing various water quality aspects.
53
the importance of real-time thermal monitoring in Notably, water temperature impacts DO saturation.
detecting thermal pollution or effluent discharges, the Higher temperatures result in lower DO saturation
system’s temporal resolution provides significant value. levels. Additionally, turbidity is a vital water quality
Continuous temperature data can also be used to support indicator, which is strongly affected by rainfall events. 46
modeling of DO dynamics or heat plume dispersion
from industrial outflows. 3.3. TDS
Temperature is an important indicator of water quality. Disposal of agricultural waste, household waste, and
The temperature differences observed can be attributed open excretions will contribute to higher turbidity
to the time of sampling, the position of the Sun, and values and increase pollutants that threaten household
the direction and shade of the Sun’s rays. The result of and irrigation use. TDS measures water pollution from
47
temperature in Banjardowo River, Semarang is shown in sewage, untreated natural sources, urban runoff, and

Volume 22 Issue 4 (2025) 9 doi: 10.36922/AJWEP025110069