INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XV October 2025 | Special Issue on Economics
Page 1336
www.rsisinternational.org
Using Satellite Remote Sensing Data in Monitoring Water Quality at
Ndakaini Reservoir Dam, Kenya
Kibetu Dickson Kinoti, Boniface Muriuki Kaburu
Department of Social Sciences, Chuka University
DOI: https://dx.doi.org/10.47772/IJRISS.2025.915EC00751
Received: 02 October 2025; Accepted: 08 October 2025; Published: 10 November 2025
ABSTRACT
Pollution from anthropogenic and natural processes threatens most fresh water resources. In Kenya, drought
and occasional floods affects water quality negatively impacting health of humans and ecosystem. Currently
the use of in situ observation methods for monitoring water quality is inadequate, periodic and at selected
sample points. The potential of remote sensing-based techniques in examining water quality in open water
bodies is explored. Aim of this study was to assess variability in Chlorophyll concentration (Chl_a), water
temperature (SST) and Suspended Sediment Concentrations (SSC) of Ndakaini Dam before and after the
destructive April-May 2018 floods. Landsat 8 OLI imagery and QGIS V 2.14 Essen geospatial software were
used. The results indicated that Chlorophyll concentration levels decreased by 0.26mg/l, water temperature
dropped by 3
0
C while Suspended Sediment increased by 2.57mg/l. Generally, analysed water quality
parameters showed a spatial and temporal variation across the April-May 2018 floods. Ndakaini dam was
found to be a low-level pollution mesotrophic reservoir.
Keywords: Ndakaini Reservoir, Remote sensing, Water Quality, real-time monitoring, Pollution
INTRODUCTION
Fresh inland water bodies are important for recreational, economic and ecological use. Globally, demand for
fresh water has escalated leading to excessive abstraction. On the other hand, the ever-growing water demand
has increasingly fueled local water crises. On the same note the quality and quantity of fresh water resources
has declined owing to population growth, migration, industrialisation and urbanization (Tessema et al., 2014;
Torbick et al., 2013; WWAP, 2015). Today, threats of pollution, land use and climate change pause serious
challenges as far as water conservation and management is concerned. Unregulated human activities have
affected water bodies resulting to seasonal water scarcity, increase in nutrients and sediment loadings
(Mushtaq & Pandey, 2014).
Surface waters are amongst the most polluted water resource from industrial, domestic and municipal sources
(Corcoran et al., 2010). Water for human consumption require high standards of quality and therefore
monitoring provides key information for detecting contaminants and assessing the suitability of available water
resources for human use (UNEP,2010; WMO,2012).
Major challenge on effective monitoring of water quality is due to lack of adequate, consistent and update
information (UNEP, 2010). Traditional insitu methods of measuring water quality tend to give accurate but
discrete measurements at selected sample points and do not cover a large water body (Abdelmalik, 2018; He et
al. 2008).
Use of satellite remote sensing data in water quality assessment hold significant potential as demonstrated by
Mathews,2011; Schaeffer et al., 2013; Saadi et al, 2014. In particular, Landsat data products have been widely
used for assessment of inland lakes water clarity and chlorophyll concentrations. Landsat’s data availability,
moderate spatial resolution and reduced temporal coverage makes it widely used (Steven et al, 2002) In Kenya,
although many studies have been done on assessment of selected Water quality variables at varying spatial
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XV October 2025 | Special Issue on Economics
Page 1337
www.rsisinternational.org
levels (Gathenya et al, 2009; Mueni, 2014) the application of satellite data has not received much attention
especially in the field of water quality assessment (Song et al,2011; Chawla et al, 2020)
There is need to explore the potential of satellite based observation in predicting water quality on domestic
water use dams in Kenya. Overall objective of this research was to assess chlorophyll concentration, water
temperature and turbidity at Ndakaini dam before and after the April-May 2018 floods using Landsat 8
Operational Land Imager sensor data.
Study area
Ndakaini dam is one of the two reservoirs supplying water to Nairobi City County. This dam supplies over
84% of water to the residents of Nairobi County. The dam is located between latitudes 00
0
48
’
& 00
0
49’South
and Longitudes 36
0
49, & 36
0
51’East on an altitude of about 2,041 Meters above sea level (figure.1) The dam
land covers 1200 acres and it has a catchment area of 75sqkm. Ndakaini dam has a storage capacity of 70
million cubic meters occupying 600 acres when full. The area has well drained, deep, dark reddish brown,
friable clay soil (Saytarkon, 2015). Ndakaini Sub location where Thika dam (Ndakaini dam) is found has a
population of 2444 persons (KNBS, 2009) and covers area of 8.5sqKm. On average the area receives 2000 to
2500 mm of rainfall per annum.
Figure.1: Location of Ndakaini Dam on South Eastern slopes of Abardares ranges (Source: Author)
MATERIALS AND METHODS
Materials
Software
ACOLITE version 3.2 software was used for radiometric calibration of Landsat 8 OLI sensor to generate
spectral radiance and reflectance values. In particular, ACOLITE was used to decode Landsat MTL text file
and build an expression to create a radiance image from the raw pixel digital numbers. SEADAS was used to
visualise derived land to water Output parameters. QGIS version 2.14 was used in delineating region of
interest and to raster clip the area of study from the two multi-temporal images.
Satellite images
Landsat 8 OLI datasets of 29
th
January 2018 and 12
th
October 2018 were used. Level 1 Landsat 8 OLI data was
downloaded from USGS Earth Explorer and later converted to tiff for input into image processing software.
Landsat 8 OLI has a temporal resolution of 16 days with spatial resolution of 15m for Panchromatic, 30m for
visible band and 100m for thermal band (USGS, 2016). In this study band 1, band 2, band 3, band 4 and band
10 respectively of Landsat 8 OLI sensor were used (table.1). Images downloaded were those with minimal
cloud cover since clouds obstruct data returning “No data” value from the ground features.
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XV October 2025 | Special Issue on Economics
Page 1338
www.rsisinternational.org
Table1. Landsat 8OLI Bands specifications as used in water quality monitoring
Band
Path/ Row Date of Acquisition Band wavelength
1
2
3
4
10
168/ 61 29-01-2018/ 12-10-2018 433nm - 453nm
168/61 29-01-2018/ 12-10-2018 450nm - 515nm
168/61 29-01-2018/ 12-10-2018 525nm - 600nm
168/61 29-01-2018 / 12-10-2018 630nm – 680nm
168/61 29-01-2018/ 12-10-2018 10300nm – 11300nm
METHODS
Landsat Data Acquisition
Landsat 8 OLI images of path 168, Row 61 were downloaded from USGS Earthexplorer website
(http//www.usgsearthexplorer). Landsat images used in this study were those with cloud cover of less than
10%. Images of 29
th
January 2018 and 12
th
October 2018 had cloud cover less than 8% and were selected to
represent “before” and “after” the April-May 2018 floods scenario.
Image pre-processing
First, radiometric correction was done on spectral bands 1, 4 and 10 by converting the digital numbers to
spectral radiance signals. This was realised by using a Chander & Markham, 2003 expression; Lλ=[(LMAXλ-
LMINλ) / (QCALMAX)] * QCAL + LMINλ]
In order to retrieve water surface reflectance values, atmospheric correction was done on the resultant image.
Then, resultant spectral radiance was converted to top of atmosphere reflectance using equations 1 and 2
obtained from USGS website (http://landsat.usgs.gov);
ρλ'=Mρ*Qcal + Aρ; Equation (1)
Thereafter, the top of atmosphere reflectance was corrected using sun angle correction equation as shown;
ρλ= ρλ'/ cos (θsz * π/180) = ρλ'/sin (θse * π/180); Equation (2)
3.2.3 Mapping Chlorophyll Concentration Levels
To map levels of chlorophyll_a concentrations across the dam, normalised digital water index (NDWI)
algorithm expression by Zhang & Hang, 2015 was used.
Chl_a = 17.878*(OLI 4 –OLI 1/ OLI4 +OLI 1) +5.636
Where; OLI 4 is red band, OLI 1 is coastal / aerosol band of land sat 8 Operational Land Imager (OLI)
sensors.
Calculating Water Surface Temperature variations
Temperature influence amount of oxygen dissolution in water. Thermal Infrared (TIRS) band 10 of land sat 8
OLI was used. Thermal Infrared raw digital numbers were converted into the top of atmosphere spectral
radiance using the radiance expression in Landsat Tools V1.0.34;
Lλ = (0.0003342 * Qcal_B10) + 0.1
To generate temperature map, top of atmosphere spectral radiance was converted into brightness temperature
using thermal constants provided in the metadata file;
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XV October 2025 | Special Issue on Economics
Page 1339
www.rsisinternational.org
Temp= 1321.0789 / (Ln (774.8853 / Rad_B10) + 1)
Created temperature maps were converted into degrees Celsius by subtracting a value 273.15 from the
resultant maps
3.2.5 Mapping Suspended Sediments
To map spatial and temporal variation in the concentrations of suspended sediments polynomial model by
Kumar, (2016) was applied to calculate the amount of suspended particles from remote sensing data;
SS= 13181* OLI 4
2
-1408.6 * OLI 4 + 44.15
Tropical State Index (TSI) estimation for Ndakaini inland Reservoir Lake
To understand the quantity of algal biomass content in Ndakaini dam, tropical state index was used. Carlson
developed an expression to map trophic state of lakes based on Secchi disk transparency, chlorophyll_a
concentration or total phosphorus content (Carlson,1977). For this research chlorophyll_a was used to calculate
the tropical state index using a formula by Pulak et al, 2016
Tropic State Index Chlorophyll =10 [6-(2.04-0.68ln(chl-a))/ Ln2]
Figure 2: Simplified flow chart of the water quality variables retrieval methodology
RESULTS AND FINDINGS
Spatial-temporal variation in Chlorophyll_a levels
Chlorophyll_a concentration levels within Ndakaini dam on 29
th
January 2018 were found to be in the range
between 5.87 Mg/l to 20.9 Mg/L with an average concentration of 10.42 Mg/L and a shown a standard
deviation of 5.25 Mg/L (fig.3). January was selected as the “dry “month image before the floods. Usually
during this month, increased evapotranspiration due low precipitation leads to reduced water quantity in the
reservoir which in turn lead to raised concentration of the chlorophyll-a.
Figure.3: Map of Chloropyhl_a Levels on 29
th
/01/2018
Selection of Spectral bands
(OLI2, OLI3, OLI4, OLI5 AND OLI10)
LANDSAT 8 Image Acquisition
Prediction of Water Quality Variables
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XV October 2025 | Special Issue on Economics
Page 1340
www.rsisinternational.org
Chlorophyll_a concentration levels in the reservoir on 12
th
October 2018 were lower compared to those
mapped on January (10.42 Mg/l to 10.16 Mg/l). The concentrations of chlorophyll_a varied from 7.94 Mg/L to
15.60 Mg/L and mean concentration of 10.16 Mg/L with standard deviation of 2.51 Mg/L as shown in figure 4.
As the rain season set in, increased overland flow into the reservoir from streams result to volumetric capacity
increasing which reduces the concentration of the chlorophyll-a within the dam.
Figure.4: Map of Chloropyhl_a Levels on 12
th
/10/2018
Dam’s Open Water Surface Temperature Dynamics
The dam’s open water surface temperature as calculated and ultimately map on 29
th
January 2018 ranged
between 21.7
0
C to 32.3
0
C with mean temperature of 27.0
0
C (Fig.5). These were higher compared to those
calculated for the month of October 2018. During this month the water levels were low and as such the open
water heating from the surface transfer a lot of direct heat into the reservoir bottom which raise the water
temperature conventionally.
Figure.5: Map of Water temperature for Ndakaini Dam on 29
th
/01/2018
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XV October 2025 | Special Issue on Economics
Page 1341
www.rsisinternational.org
October showed lower water surface temperatures being after the April-May 2018 long rains. The mapped
temperature ranged from 19.9
0
C to 28.03
0
C. The month’s average temperature was 24.0
0
C. This was attributed
to the increased reservoir water levels from stream inputs draining into dam (Fig.6). After the floods, the water
levels in the dam increased which led to even distribution of water heat by currents translating to lowering of
the open surface water temperature.
Figure.6: Map showing Water temperature for Ndakaini Reservoir on 12
th
/10/2018
Suspended Sediments Concentration variations
Suspended Sediment loadings across the entire dam were map during the months of January and October.
Average amount of suspended sediment concentrations as calculated from Landsat 8 OLI bands on 29
th
January and 12
th
October ranged between 45.84 Mg/L in January to 48.41 Mg/L on October 2018 (Fig.7). The
increase of 2.57 Mg/L sediment concentration between January and October was linked to surface run-offs and
increase stream input. The observed variation is associated with sediment load distribution within the reservoir
by inflowing stream discharge which brought more sediment in as much as it aided in the spreading it within
the reservoir.
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XV October 2025 | Special Issue on Economics
Page 1342
www.rsisinternational.org
Figure.7: Map of Suspended Sediment Loads on 29
th
/01 2018 and 12
th
/10/2018
DISCUSSION AND CONCLUSION
The observed spatial and temporal variation in chlorophyll a concentration, surface water temperature and
suspended solid components during and after the April 2018 floods was attributed to dynamics in dam nutrient
inputs and water volume. The tropic state index value of this lake indicates Ndakaini dam is mesotrophic lake
water with low level of pollution. The lake’s capacity varies from 457.02 acres before the floods (low water
volume) to 574.89 acres after the flood (High water volume) as shown in figure 8 below. This also explains the
observed low Chlorophyll a concentration in October compared to January due to increased water input.
Figure. 8: (A) Before the flood’s lake extent (457.02Acres)
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XV October 2025 | Special Issue on Economics
Page 1343
www.rsisinternational.org
Figure.8: (B) After the flood’s lake extent (574.89Acres)
This study advocates the use of remote sensing data for long-term monitoring of water quality conditions as
well as in predicting water quality variables for open surface water bodies which are inaccessible. Furthermore,
there is a need to integrate Geospatial tools with insitu measurements to strengthen the current water
monitoring and assessment methods if Kenya’s surface water resources are to help achieve the Vision 2030
and SDG no.6.
REFERENCES
1. Abdelmalik, K.W. (2018). Role of statistical remote sensing for Inland water quality parameters
prediction. The Egyptian Journal of Remote Sensing and Space Sciences 21:193-200
2. Carlson,R.E(1977). A trophic state index for lakes. Limnology of. Oceanography.22:361-369
3. Chander, G. and Markham, B., 2003, Revised Landsat-5 TM radiometric calibration procedures and
postcalibration dynamic ranges. IEEE Transactions on Geoscience and Remote Sensing,41(11),
2674-2678.
4. Corcoran, E., C. Nellemann, E. Baker, R. Bos, D. Osborn, H. Savelli (eds). (2010). Sick Water? The
central role of wastewater management in sustainable development. A Rapid Response Assessment.
United Nations Environment Programme, UN-HABITAT
5. Gathenya, J. M., Thiong’o G. T. and Mwangi, J. K. (2009). Hydrological modelling and water
Quality Assessment of Sasumua watershed
6. He W, Chen S, Liu X, Chen,J.(2008). Water quality monitoring in a slightly-polluted nland water
body through remote sensing – case study of the Guanting reservoir in Beijing, China. Front Environ
Sci Eng China. 2:163–171.
7. Kumar, A., Equeenuddin, Sk. Md., Mishra, R. D, and Acharya B., C., 2016. Remote Monitoring of
Sedimen Dynamics In Coastal Lagoon: Long Term Spatio-Temporal Variability of Suspended
Sediment In Chilika. Journal of Estaurione, Coastal and Shelf Science, 155-172
8. Matthews, M.W.(2011). A current review of empirical procedures of remote sensing in inland and
near-coastal transitional waters. International Journal of Remote Sensing, 32 (21), 6855-6899.
9. Mueni,P.J (2016). Climate change impacts on water resources over the upper Tana catchment of
Kenya. Unpublished Msc thesis, University of Nairobi
10. Musau, J., Sang, J., Gathenya, J., & Luedeling, E. (2015). Hydrological responses to climate change
in Mt. Elgon watersheds. Journal of Hydrology: Regional Studies.3: 233-246
11. Mushtaq, F, Pandey, A.C. (2014). Assessment of land use/land cover dynamics vis-a-vis
hydrometeorological variability in Wullar Lake environs Kashmir Valley, India using multitemporal
satellite data. Arabian Journal of Geosciences. 7:4707–4715
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XV October 2025 | Special Issue on Economics
Page 1344
www.rsisinternational.org
12. Pulak, P, Sourabh, K., Raman, K., Sanjeev, K. and Sangram, K. (2016) Estimation of Chlorophyll‐a
Concentration and Trophic States for an Inland Lake from Landsat 8 OLIData: A Case of Nalban
Lake of East Kolkata Wetland, India. Non-peer reviewed Article
13. Saadi, A.M., Yousry, M.M., Jahin, H.S., (2014). Statistical estimation of Rosetta branch water
quality using multi-spectral data. Water Science Journal, 28: 18–30
14. Saytarkon, D.O (2015). The social and environmental impacts of Ndakaini dam on the catchment
community, Kenya. Unpublished MA Thesis, University of Nairobi
15. Schaeffer, B. A., Schaeffer, K. G., Keith, D., Lunetta, R. S., Conmy, R. and Gould, R. W. (2013).
Barriers to adopting satellite remote sensing for water quality management. International Journal of
Remote Sensing, 3421(21), 7534–7544.
16. Steven, M. K, Patrick, L. B, Leif, G. O, Marvin, E. B.(2002) A procedure for regional lake water
clarity assessment using Landsat multispectral data. Remote Sensing of Environment, 82:38-47
17. Tessema, Mohammed, B. T. (2014). Assessment of Physico-chemical Water Quality of Bira Dam,
Bati Wereda. Journal of Aquactic Resources Development, 5(6),1-4.
18. Torbick N, Hession S, Hagen S, Wiangwang N, Becker B, Qi J. (2013). Mapping inland lake water
quality across the Lower Peninsula of Michigan using Landsat TM imagery. Int J Remote Sens.
34:7607–7624.
19. UNEP. (2010). Clearing the waters. A focus of water quality solutions. Accessed from
http://www.unep.org on 16
th
October 2018
20. USGS (2016). Provisional Landsat 8 Surface Reflectance Product Guide. Retrieved on from
https://landsat.usgs.gov/documents/provisional_l8sr_product_guide.pdf on 23
rd
August 2018
21. WMO. (2012). Technical material for water resources assessment. Technical Report Series. WMO
1095. Accessed from http://www.wmo.int/publications on 5th September 2018.
22. WWAP (United Nations World Water Assessment Programme) ( 2015). The United Nations World
Water Development Report 2015: Water for a Sustainable World. Paris, UNESCO
23. Zhang, C. and Hang, M. (2015). Mapping Chlorophyll-a Concentration in Laizhou Bay Using
Landsat 8 OLI Data. E-proceedings of the 36thIAHR World Congress 28 June- 3 July, 2015. The
Hague, Netherlands
24. Song, K.; Wang, Z.; Blackwell, J.; Zhang, B.; Li, F.; Jiang, G (2011). Water quality monitoring
using Landsat Thematic Mapper data with empirical algorithms in Chagan Lake, China. Journal of
Applied Remote Sensing
25. Chawla, I.; Karthikeyan, L.; Mishra, A (2020). A review of remote sensing applications for water
security: Quantity, quality, and extremes. Journal of Hydrology,585, 124826.