SN Applied Sciences

, 2:85 | Cite as

Development of dual water supply using rooftop rainwater harvesting and groundwater systems

  • Siti Nazahiyah RahmatEmail author
  • Adel Ali Saeed Al-Gheethi
  • Syafiqa Ayob
  • Fitryaliah Mohd Shahli
Research Article
Part of the following topical collections:
  1. Engineering: Advances in Civil Engineering: Towards a Sustainable Future


The aim of this study was to gain a better understanding on dual water supply system consisting of rainwater and groundwater. The pilot study of the designed system was carried out at the Universiti Tun Hussein Onn Malaysia, Batu Pahat, Johor. The first part of the analysis was carried out to assess the rainwater harvesting system efficiency. The monthly results on total quantities of the collected rainfall in storage tank was 48.97 m3 while the actual rainfall calculated was 56.04 m3 with 75% of collection efficiency. The rainfall-storage rating curve was then plotted using 150 rain events data versus the volume of harvested rainwater collected during the study period. The second part of the analysis was to determine the total average pumping rate by conducting a step drawdown test, which gave Q = 39.5 m3 / h. For the daily discharge rate, the well was capable to supply water approximately 1.69 m3. In the current study, the daily water demand calculated was 0.59 m3, which gave the total volume per month approximately from 16.5 to 18.3 m3. Based on the performance of the system, most of the days rainwater could not meet the water demand, thus have to be supported by the groundwater.


Conservation Groundwater Rainwater harvesting system Water supply 

1 Introduction

Malaysia is known to have abundant surface water due high average annual rainfall between 2420 and 3830 mm. In the extreme cases, Malaysia receives heavy rainfall between November and February as much as 600 mm in 24 h. This statistics shows that the annual and monthly rainfall in Malaysia are quite large [1]. However, in some cities water shortage or scarcity continues to persist due to the increasing in the annual population growth of 1.5% and industrial activity. In addition, the water tariffs in Malaysia is extremely cheap compared to Indonesia and Singapore, therefore, the average daily water consumption of Malaysians is between 220 and 250 L, which is much more than that recommended by World Health Organization of 165 L/day. An effective method of water supply and storage system should be taken seriously to avoid the demand problem, especially in drought season. Public and private water user agencies are encouraged to shift to integrated management of water. The ability to be close to being self-sufficient on water resources would render the building to be sustainable in terms of water resources. In order to be self-sufficient, alternative water resources should be utilized instead of relying on public water supply. RWH and groundwater abstraction are among the solutions to the problem in the areas that have inadequate water resources or face long-term water supply disruptions as well as to complement conventional systems for non-potable use, such as toilet flushing, clothes washing, watering plants, irrigation and for potable uses [2, 3, 4, 5, 6].

RWH is said to have very good potential as an alternative water supply in Malaysia. RWH was introduced after the 1998 drought by Ministry of Housing and Local Government. Few guidelines concerning RWH in Malaysia have been developed [7]. According to the recent studies which have been carried out on the rainwater quality in Malaysia, RWH has the potential to reduce cost and the usage of treated water as well as can be considered as one of the alternative for the water resources [8, 9]. Although RWH is gaining much interest, the implementation has been limited due to several disadvantages such as seasonal variation and related cost [7, 10, 11, 12, 13]. Recently, Hafizi Md Lani et al. [13] concluded that Malaysia is well positioned to harvest rainwater and to be a saviour for the water shortage, but in some places or event, it cannot completely dependable source of water supply due to uneven rainfall distributions. As to support the RWH system, other natural resources can be considered. For instance, by making a dual water supply system consisting of rainwater and groundwater.

Groundwater is an alternative water source in the areas with a limited surface water resources. In Malaysia, the utilization of groundwater is very low compared to other countries such as Denmark, Austria, Thailand, China, USA, due to failure to recognize the vast potential of the invisible groundwater resource [14]. Hence, to test the level of usability of groundwater a comprehensive study needs to be conducted. In addition, rainwater and groundwater are two interdependent water sources. However, people who harvest rainwater generate an additional negative impact on groundwater users. In other words, although the technology of RWH system may allow us to collect the runoffs, it also causes decreasing amount of water available to replenish the aquifer. Therefore, the motivation of this study was to develop a water supply system that integrates both rooftop RWH system and groundwater abstraction as water supply alternatives to fulfil the water demand while adopting the concept of sustainability and conservation on water resources.

2 Methodology

2.1 Study area

The pilot study of the current designed system was carried out at the Universiti Tun Hussein Onn Malaysia (UTHM), Batu Pahat, Johor, Malaysia (Fig. 1) (1°51′N, 102°56′E). The city is located 239 km to the south of Kuala Lumpur. The mean daily temperature is about 27 °C and the average monthly rainfall data ranged from 75 to 230 mm. The rainfall is the lowest in January, with an average of 75 mm, while the highest amount of precipitation of 230 mm occurs in November. The study was conducted between December 2018 and May 2019. In the current study, the water demand and rainwater harvesting (RWH) system were first to be designed. For groundwater exploration, the work started with the pumping work to determine the hydrogeology properties.
Fig. 1

Location of Universiti Tun Hussein Onn Malaysia in Johor, Malaysia

2.2 Rainwater harvesting system

The design of the RWH system was fully referred to Department of Irrigation and Drainage (DID) [15] known as the Urban Storm Water Management Manual for Malaysia. The main components of the system were rainwater tank, conveyance and rooftop. The tank size for Malaysia regardless of location was 1 m3 for roof area of 100 m2. It is equivalent to store 10 mm of rainfall with 100 m2 of roof area [16]. It can be estimated using Eq. 1.
$$S_{t} = 0.01Ar$$
where \(S_{t}\) is the tank size (m3) and Ar is the rooftop catchment area (m2).
The flow during the rain events was delivered to the tank via gutters and downpipe, and temporarily stored before use. The gutter width and diameter of downpipe were determined using Table 3 in Appendix. The width of gutter is always half of the gutter depth [15]. The complementary devices for quality control were debris screens and first flush diverters. The diverter was used to prevent the first flush of water from entering the tank and diverts the water flow to the tank. For each 60 m2 of roof, it was assumed that 1.0 mm was used as first flush. Hence, Eqs. 2 and 3 were used to get the total volume of first flush.
$$V = L \times b \times h$$
$$P_{L} = V/\pi r^{2}$$
where V is the required volume of diverted water \(\left( {m^{3} } \right)\), L is the roof length (m), b is the roof width (m), h is the first flush depth, \(P_{L}\) is the pipe length (m) and \(r\) is the diameter of the pipe (m). Once the design work was completed, the RWH system was installed.

2.3 Water demand

The rainwater demand depends on the number of people using the water, average consumption per person, and the range of uses by the consumer. The water demand for the rainwater influences the effectiveness and availability of designed rainwater storage tank. Rainwater demand for domestic application recommended by DID [16] was used as a guideline in designing process. In the current study, for indoor usage with average of four households, a single flush toilet was chosen with average total water used was 0.12 m3 per day. Meanwhile, for a twin tub washing machine, the average total water used was 0.4 m3 per day. Estimated average total for dishwasher and general cleaning were 0.3 and 0.2 m3 per day, respectively. For outdoor uses, total average for drip system, sprinkler, and hosing paths were estimated to be 0.2 m3 per day. Therefore, the total daily water demand was 0.59 m3.

2.4 Rainfall analysis

Rainfall data was measured and digitally recorded using HOBO tipping bucket rain gauge. The rain gauge was installed on a levelled platform on a tower to make sure there was no surrounding obstacle. A rating curve was established by making a number of concurrent observations of rainfall and storage over a period of time, covering the expected range of stages at the rainfall gauging section. The physical change to the rainfall data alters this relation and these changes were accounted for in the derivation of a storage from a time series of storage data. The idea of rainfall-storage rating curve was adopted from the conventional stage–discharge rating curve where the data are plotted versus the concurrent stage to define the rating curve for the stream [17].

To measure the effectiveness of the RWH system, the observed and actual amount of harvested rainwater were identified. The observed harvested rainwater were monitored daily at the study site. While, the potential of water harvesting was estimated using Eq. 4 [18]. However, due to real world system inefficiencies such as roofing material, a slight reduction in water collection needs to be considered and is calculated using Eq. 5. This reduction is called collection efficiency or runoff coefficient. It indicates the fraction of rainwater that can be collected from roof. Thus, the analysis between potential and actual amount in water collection provides the actual value on collection efficiencies of the installed system.
$$1\,{\text{L}}\; of\;rainwater = 1\,{\text{m}}^{3}\; of \;catchment\; area \times 1\,{\text{mm}} \;of\; rain$$
$$Potential \,water\, collection\,({\text{L}}) = Roof\, Catchment \,area\,({\text{m}}^{2} ) \times Annual\, precipitation\,({\text{mm}})$$
$$Actual\; water \;collection\,\left( {\text{L}} \right) = Roof \;catchment\; area\,({\text{m}}^{2} ) \times Annual\; precipitation\,({\text{mm}}) \times Collection \;efficiency$$

2.5 Groundwater pumping test

Constant rate test was carried out to estimate the well yield. The well was pumped for 100 min at a constant rate and the drawdown was measured at frequent intervals. Once the pump was stopped, the recovery data was also recorded similar as the drawdown time scale [19, 20]. The rate of the groundwater discharge from the well to the storage tank was calculated using Eq 6.
$$Q = \forall \times \frac{24}{t}$$
where \(\forall\) is the volume per hour and t is the recovery hour.

2.6 Dual water supply system

In the current study, rainwater was used in preference to groundwater and the system was equipped with automatic devices. The switching device used floats to automatically detect when sufficient rainwater was available for use in the building. When rainwater levels were low, the device automatically changed back to groundwater so there was no an uninterrupted supply of water to the building.

For rooftop RWH, the size of the rooftop was 129.5 m2, hence, 1 m3 storage tank was proposed. This rainwater slim tank was made of high-density polyethylene and suitable for storage of slight acidic rainwater. The size of the gutter used was 7.8 m2 and 150 mm diameter of circular downpipe was chosen in the current study. For first flush diverter, the required volume of diverted water was 252 L. Two diverters were used to cater the flows from the rooftop. Figure 2 illustrates the schematic diagram and the water flow of the full developed system. It consisted of 12 main components as described in Table 1. For harvested rainwater, the rainwater was collected using the roof catchment and directly transferred to its storage tank. However, to prevent large debris from entering the storage tank, the rainwater was drained through the first flush diverter. Meanwhile, for groundwater, it was extracted by using the mechanical pump from the tube well. The extracted groundwater was pumped and stored in the storage tank. The final component in this system was a circular polyethylene tank with 2 m3 capacity. The tank was used as the distribution tank where the harvested rainwater and extracted groundwater are distributed to the users.
Fig. 2

Schematic diagram of the developed system

Table 1

Main components of the installed systems





Roof catchment

Length = 14 m and Width = 9 m



Ø = 0.3 m


Overflow pipe

Ø = 0.3 m



Capacity = 1000 L



Length = 14 m



Length = 2 m, Ø = 0.3 m



Capacity = 2000 L



Capacity = 1000 L


Electrical pump





11 and 12

Groundwater pipe

Length = 20 m

3 Results and discussion

3.1 Rainfall-storage rating curve

To get the correlation between rainfall and water collection, a rainfall-storage rating curve was developed. The rating curve was developed using 150 rain events data and the volume of harvested rainwater collected within the study period (Fig. 3). A linear trendline, \(volume \,in\, storage\, tank \left( {{\text{m}}^{3} } \right) \,is\, 0.0682 \times Rainfall\, depth \left( {\text{mm}} \right)\) was formed to show the correlation between the depth of rainfall and the volume of harvested rainwater. The correlation coefficient, r, was determined using Pearson’s method to measure the statistical relationship between two continuous variables. The value of R is 0.9937, which gave a strong positive correlation and the value of the coefficient of determination, R2, is 0.9875.
Fig. 3

Rainfall-storage rating curve

3.2 Constant rate test

A constant rate test was carried out using an electrical pump at the study area. The pump was switched on for 100 min and the drawdown data was recorded. Once the pump was stopped, the recovery data was measured similar as the drawdown time scale. Figure 4 illustrates the results of the drawdown and recovery data from the pumping test that was carried out. The constant pumping test reveals the maximum rate of groundwater withdrawal to avoid over exploitation. The recorded discharge rate for 100 min while conducting the test was 3.375 × 10−3 m3 per min, hence, the volume of extracted groundwater was determined. From the constant rate test conducted, the drawdown and recovery period recorded were 1.67 h and 1.17 h, respectively. Hence, the daily discharge rate of the groundwater from the well to the storage tank was calculated using Eq. 5 and gave 1.69 m3.
Fig. 4

Drawdown and recovery data for constant rate test

3.3 Water demand and water usage

The calculated daily water demand as shown earlier was 0.59 m3 which gave the total volume per month approximately from 16.5 to 18.3 m3. The water demand, monthly rainwater and groundwater used are depicted in Fig. 5. In December 2018, total monthly depth of rainfall was approximately 100.4 mm. The total volume of harvested rainwater was 6 m3 and complemented with 12.3 m3 of groundwater (Fig. 5a). In January 2019, the total rainfall depth observed was much higher than the previous month with approximately 204.4 mm. However, due to the limitation of the storage tank, only 6.2 m3 of rainwater were captured. Figure 5b shows that January recorded the groundwater usage of 12.1 m3. In February 2019, the total rainfall depth captured was 93.8 mm, which was the lowest rainfall depth during the study period. The amount of harvested rainwater and groundwater used were 5.6 m3 and 10.9 m3, respectively Fig. 5c.
Fig. 5

Volume of water needed from harvested rainwater and groundwater to meet water demand

Meanwhile, March recorded the least amount of monthly groundwater usage of 6.2 m3 (Fig. 5d). The recorded rainfall depth was 197 mm with the total amount of rainwater collected in the storage tank of 11.5 m3. Increase in rainwater storage allows to reduce groundwater pumping by a more or less equivalent amount. Almost similar results found in April 2019. The amount of harvested rainwater and groundwater used were 11 m3 and 6.7 m3, respectively. In May 2019, the total volume of harvested rainwater was 8.6 m3 and complemented with 9.7 m3 of groundwater. Overall, it can be seen that most of the days rainwater could not meet the water demand thus have to be supported by the groundwater. However, in doing so, not only the rainwater could be captured and reused but there is no over-exploitation of groundwater issues occur in the future.

From the results discussed earlier, January recorded the highest monthly groundwater usage with 12.3 m3. Meanwhile, March recorded the least amount of monthly groundwater usage of 6.2 m3. From the constant rate test conducted, the daily discharge rate of the groundwater was 1.69 m3. Therefore, the pump was capable to extract approximately 50.7 m3 of groundwater per month. Overall, it can be concluded that the groundwater was able to be used as supplement water to the system without any over exploitation issues since the required groundwater volume to meet the monthly water demand ranged from 6.2 to 12.3 m3.

3.4 Observed and actual amount of rainwater collection

During the study period from December 2018 to May 2019, the monthly rainfall depth recorded at the study area were as follows: 100.4 mm in December 2018, 204.4 mm in January 2019, 93.8 mm in February 2019, 197 mm in March 2019, 279 mm in April 2019 and 318.6 mm in May 2019. In order to determine the effectiveness of the installed RWH system, actual volume of rainwater collected was identified. Equation 4 was used to identify the actual amount of rainfall that can be collected. The harvested rainwater were monitored daily at the study site. In the present study, the runoff coefficient was 0.75 since it gave the least differences volume for both observed and actual rainwater collected. This result was comparable to that suggested by NAHRIM (2014) which for corrugated metal sheet roofs, the runoff coefficients ranged between 0.7 and 0.9. In Seoul, South Korea, Mun and Han [21] estimated the collection efficiency of rainwater from a building rooftop of 0.8. A runoff coefficient of 0.85 was employed in a study carried out by Adugna et al. [22] to account for evaporation loss and possible first flush diversion.

Overall, the actual volume of calculated harvested rainwater ranged from 6.4 to 12.5 m3, while the observed volume of harvested rainwater in the storage tank ranged from 5.6 to 11.5 m3 (Table 2). The differences indicated the total losses of the observed volume of harvested rainwater, which may due to overflow. The system can actually harvest more rainwater as the study area received massive rainfall in certain months, however, the tank size was limited to 1 m3. For instance, in December 2018, total monthly depth of rainfall was approximately 100.4 mm. The total volume of harvested rainwater was 6 m3. In January 2019, the total rainfall depth observed was much higher than the previous month with approximately 204.4 mm. However, due to the limitation of the storage tank, only 6.2 m3 of rainwater were captured. A more intensive monitoring program is recommended and preferably include various storm sizes to determine the optimal size of the tank size. It has been proved by Nasif and Roslan [23] in their conclusion of the study on the effect of varying roof runoff coefficient values and tank sizes.
Table 2

Differences between observed and actual amount of harvested rainwater


Rainfall (mm)

Volume of harvested rainwater (m3)


































4 Conclusions

In Malaysia, exploring alternative water sources and conserving natural resources have always been important issues in the context of water resources management. Therefore, the main aim of this study was to adopting the RWH system and groundwater abstraction as part of the conservation of natural water resources. This has also been investigated in some countries such as in France [24, 25] and Vietnam [26] and successfully implemented. To determine the collection efficiencies of RWH system, a comparison between the actual rainwater that can be harvested and the observed harvested rainwater was made. It can be concluded that 0.75 was the most accurate runoff coefficient since it gave the least differences volume for both observed and actual rainwater collected. To assess the yield of the tube well, a constant rate pumping test was carried out and the safest discharge rate was determined. Overall, by implementing this system, it can be observed that most of the months, rainwater could not meet the water demand thus have to be supported by the groundwater.

Overall, this study provides a significant reference for sustainable water management and this approach can be efficiently applied especially for rural areas, where there is no piped water supply. Water quality for both rainwater and groundwater warrant further investigations as the current study only focuses on the water quantity.



Siti Nazahiyah Rahmat was supported by a Fundamental Research Grant Scheme (FRGS) granted by the Ministry of Higher Education, Malaysia, Grant Number: 1523.

Compliance with ethical standards

Conflict of interest

The authors declare that they have no conflict of interest.


  1. 1.
    Chan NW (2015) Impacts of disasters and disaster risk management in Malaysia: the case of floods. In: Aldrich DP, Oum S, Sawada Y (eds) Resilience and recovery in Asian disasters: community ties, market mechanisms, and governance. Springer, Tokyo, pp 239–266CrossRefGoogle Scholar
  2. 2.
    Lade O, Oloke D (2015) Modelling rainwater system harvesting in Ibadan, Nigeria: application to a residential apartment. Am J Civil Eng Archit 3(3):86–100Google Scholar
  3. 3.
    Kumar MK (2015) Design of rainwater harvesting system at Shilpa Hostel in JNTUA College of Engineering Ananthapuramu: a case study from Southern India. Int J Eng Res Dev 11(12):19–29Google Scholar
  4. 4.
    Okoye CO, Solyalı O, Akıntuğ B (2015) Optimal sizing of storage tanks in domestic rainwater harvesting systems: a linear programming approach resources. Conserv Recycl 104:131–140CrossRefGoogle Scholar
  5. 5.
    Sanches Fernandes LF, Terêncio DPS, Pacheco FAL (2015) Rainwater harvesting systems for low demanding applications. Sci Total Environ 529:91–100CrossRefGoogle Scholar
  6. 6.
    Vialle C, Busset G, Tanfin L, Montrejaud-Vignoles M, Huau MC, Sablayrolles C (2015) Environmental analysis of a domestic rainwater harvesting system: a case study in France. Resour Conserv Recycl 102:178–184CrossRefGoogle Scholar
  7. 7.
    Lee KE, Mokhtar M, Mohd Hanafiah M, Abdul Halim A, Badusah J (2016) Rainwater harvesting as an alternative water resource in Malaysia: potential, policies and development. J Clean Prod 126:218–222CrossRefGoogle Scholar
  8. 8.
    Sultana N, Akib S, Aqeel Ashraf M, Roseli Zainal Abidin M (2016) Quality assessment of harvested rainwater from green roofs under tropical climate. Desalin Water Treat 57(1):75–82Google Scholar
  9. 9.
    Leong JYC, Chong MN, Poh PE, Hermawan A, Talei A (2017) Longitudinal assessment of rainwater quality under tropical climatic conditions in enabling effective rainwater harvesting and reuse schemes. J Clean Prod 143:64–75CrossRefGoogle Scholar
  10. 10.
    Che-Ani A, Shaari N, Sairi A, Zain M, Tahir M (2009) Rainwater harvesting as an alternative water supply in the future. Eur J Sci Res 34(1):132–140CrossRefGoogle Scholar
  11. 11.
    Traboulsi H, Traboulsi M (2017) Rooftop level rainwater harvesting system. Appl Water Sci 7(2):769–775CrossRefGoogle Scholar
  12. 12.
    Ayob S, Rahmat SN (2017) Rainwater harvesting (RWH) and groundwater potential as alternatives water resources in Malaysia: a review. MATEC Web Conf 103:04020CrossRefGoogle Scholar
  13. 13.
    Hafizi Md Lani N, Yusop Z, Syafiuddin A (2018) A review of rainwater harvesting in Malaysia: prospects and challenges. Water 10(4):1–21CrossRefGoogle Scholar
  14. 14.
    Taib MB (2019) Groundwater overview in Malaysia. In: National groundwater conference 2–3 July 2019Google Scholar
  15. 15.
    Department of Drainage and Irrigation, Malaysia (2012) Urban stomwater management manual for Malaysia. MSMA 2nd ednGoogle Scholar
  16. 16.
    Department of Irrigation and Drainage Rainwater (2009) Harvesting guidebook planning and design, Malaysia, 2009Google Scholar
  17. 17.
    Muzzammil M, Alam J, Zakwan M (2018) A spreadsheet approach for prediction of rating curve parameters. In: Singh V, Yadav S, Yadava R (eds) Hydrologic modeling. Water science and technology library, vol 81. Springer, Singapore, pp 525–533Google Scholar
  18. 18.
    Mourad K, Berndtsson R (2011) Potential water saving from rainwater harvesting in Syria. Vatten 67:113–117Google Scholar
  19. 19.
    Wismer L, Vergain J, Nydegger T (2011) Practical guidelines for test pumping in water wells. SwitzerlandGoogle Scholar
  20. 20.
    Macdonald A, Davies J (2001). Simple methods for assessing groundwater resources in low permeability areas of Africa groundwater systems and water quality commissioned report CR/01/168N2002Google Scholar
  21. 21.
    Mun JS, Han MY (2012) Design and operational parameters of a rooftop rainwater harvesting system: definition, sensitivity and verification. J Environ Manag 93(1):147–153CrossRefGoogle Scholar
  22. 22.
    Adugna D, Jensen BM, Lemma B, Gebrie SG (2018) Assessing the potential for rooftop rainwater harvesting from large public institutions. Int J Environ Res Public Health 15(2):336CrossRefGoogle Scholar
  23. 23.
    Nasif MS, Roslan R (2016) Effect of varying roof run-off coefficient values and tank size on rainwater harvesting system’s water savings in Malaysia. ARPN J Eng Appl Sci 11(22):12937–12941Google Scholar
  24. 24.
    Soubeyran R, Tidball M, Tomini A, Erdlenbruch K (2015) Rainwater harvesting and groundwater conservation: when endogenous heterogeneity matters. Environ Resour Econ 62(1):19–34CrossRefGoogle Scholar
  25. 25.
    Stahn H, Tomini A (2017) On conjunctive management of groundwater and rainwater. Resour Energy Econ 49:186–200CrossRefGoogle Scholar
  26. 26.
    Nguyen DC, Han MY (2014) Design of dual water supply system using rainwater and groundwater at arsenic contaminated area in Vietnam. J Water Suppl Res Technol-AQUA 63(7):578–585CrossRefGoogle Scholar

Copyright information

© Springer Nature Switzerland AG 2019

Authors and Affiliations

  1. 1.Department of Water and Environmental Engineering, Faculty of Civil and Environmental EngineeringUniversiti Tun Hussein Onn Malaysia (UTHM)Batu PahatMalaysia

Personalised recommendations