**4. Requirements and considerations for rainfall simulator**

The characteristics of raindrop is also important for storm-water management purpose particularly in relation to understanding runoff process [26, 27]. Rainfall simulation should exemplify the following fundamental characteristic of the natural rain [7, 28].

i.Drop size distribution

ii.Terminal velocity

iii.Distribution uniformity

iv.The rainfall intensity and

v.Kinetic energy

#### **4.1 Drop size distribution**

One of the basic natural rainfall characteristics that are considered is it drop size which ranges between 0.5 and 5 mm [3]. The measurement of rain droplets sizes has been studied using various approaches [28], but there is no established standard for obtaining raindrop diameter size [3]. Basically, there are two methods used for determining drop size; manual and automatic raindrop measurement [26].

The manual measurement techniques of drop size distribution includes; stain, flour pellet, oil immersion and photographic methods while automated raindrop measurement techniques include; impact disdrometers (acoustic and displacement); optical disdrometer (optical image and optical scattering). **Figure 6** presents a tree of the drop size distribution classification. Drop size can be determined using Eq. (1):

$$D\_r = \sqrt[3]{\frac{6}{\pi} W} \tag{1}$$

**99**

*A Revisit of Rainfall Simulator as a Potential Tool for Hydrological Research*

A natural raindrop from greater height tends to reach terminal velocity before impact. This impact produces several effects on soil disintegration and infiltration. This is important particularly for studying soil erosion challenges where drops should reach their terminal velocity before impact [28]. This rainfall characteristic highly depends on the height of the simulator [3]. When the downward gravitational forces acting on the rainfall are cancelled out by the drag acting on the drop, the terminal velocity is achieved. Terminal velocity have been measured by many researchers using electronic devices to estimate the time for drops to pass consecutive point via photograph during fall [1, 7, 22, 28, 29] by stopwatch, timing the individual fall from a known height or simple computation [28, 30]. Computation of velocity of drop reaching the ground at an angle in natural precipitation (storm) with wind conditions 3. Simulated rainfall if well done can attain up to 94% of the terminal velocity of the natural rainfall [3, 31].

One of the major ways to assess rainfall simulators is by the simulated rain intensity which the means by which other rainfall characteristics are defined, especially the rain impact kinetic energy [3]. Another characteristic that correlated with intensity is the drop size distribution [11]. The method to control rain intensity varies in rainfall simulator. But it is quite a difficult task most especially using drop forming simulator because it involved the manual movement of the frame [3, 11, 32]. In the case of pressurised nozzle simulators, intensity and drop diameter are control the varying the pressure [3] or by introducing a body in a swinging or

Kinetic energy of rainfall is the degree by which the energy of the rain is measured. It is the major factor in soil detachment process. The energy of the rain is

*DOI: http://dx.doi.org/10.5772/intechopen.93532*

**4.2 Terminal velocity**

*Classification of drop size distribution methods.*

**Figure 6.**

**4.3 Rain intensity**

**4.4 Kinetic energy**

rotating motion under the nozzle [6, 14].

where *W* is the weight of the formed.

*A Revisit of Rainfall Simulator as a Potential Tool for Hydrological Research DOI: http://dx.doi.org/10.5772/intechopen.93532*

*Agrometeorology*

was not recycled.

natural rain [7, 28].

i.Drop size distribution

iii.Distribution uniformity

iv.The rainfall intensity and

where *W* is the weight of the formed.

ii.Terminal velocity

v.Kinetic energy

**4.1 Drop size distribution**

performance. The merit of these simulators is that it can be used to study field parameters required for hydrologic modelling on any surface including the ones covered with vegetation. But they were limited by problem of natural rainfall which resulted to dismantling the setup when experiment schedule was not over and water

i.It is faster in data collection without waiting for the natural rain.

ii.With rainfall simulator, you can work with controlled rain, thereby,

out in 2010 using laboratory simulator [25] were pointed out as:

**4. Requirements and considerations for rainfall simulator**

From the numerous studies carried out on the simulation of rainfall both in field and laboratory experiment, two merits of rainfall simulator in a research carried

eliminating the erratic and unpredictable changeability of natural rainfall.

The characteristics of raindrop is also important for storm-water management purpose particularly in relation to understanding runoff process [26, 27]. Rainfall simulation should exemplify the following fundamental characteristic of the

One of the basic natural rainfall characteristics that are considered is it drop size which ranges between 0.5 and 5 mm [3]. The measurement of rain droplets sizes has been studied using various approaches [28], but there is no established standard for obtaining raindrop diameter size [3]. Basically, there are two methods used for determining drop size; manual and automatic raindrop measurement [26]. The manual measurement techniques of drop size distribution includes; stain, flour pellet, oil immersion and photographic methods while automated raindrop measurement techniques include; impact disdrometers (acoustic and displacement); optical disdrometer (optical image and optical scattering). **Figure 6** presents a tree of the drop size distribution classification. Drop size can be determined

> <sup>3</sup> <sup>6</sup> *D W <sup>r</sup>* π

<sup>=</sup> (1)

**98**

using Eq. (1):

*Classification of drop size distribution methods.*

#### **4.2 Terminal velocity**

A natural raindrop from greater height tends to reach terminal velocity before impact. This impact produces several effects on soil disintegration and infiltration. This is important particularly for studying soil erosion challenges where drops should reach their terminal velocity before impact [28]. This rainfall characteristic highly depends on the height of the simulator [3]. When the downward gravitational forces acting on the rainfall are cancelled out by the drag acting on the drop, the terminal velocity is achieved. Terminal velocity have been measured by many researchers using electronic devices to estimate the time for drops to pass consecutive point via photograph during fall [1, 7, 22, 28, 29] by stopwatch, timing the individual fall from a known height or simple computation [28, 30]. Computation of velocity of drop reaching the ground at an angle in natural precipitation (storm) with wind conditions 3. Simulated rainfall if well done can attain up to 94% of the terminal velocity of the natural rainfall [3, 31].

#### **4.3 Rain intensity**

One of the major ways to assess rainfall simulators is by the simulated rain intensity which the means by which other rainfall characteristics are defined, especially the rain impact kinetic energy [3]. Another characteristic that correlated with intensity is the drop size distribution [11]. The method to control rain intensity varies in rainfall simulator. But it is quite a difficult task most especially using drop forming simulator because it involved the manual movement of the frame [3, 11, 32]. In the case of pressurised nozzle simulators, intensity and drop diameter are control the varying the pressure [3] or by introducing a body in a swinging or rotating motion under the nozzle [6, 14].

#### **4.4 Kinetic energy**

Kinetic energy of rainfall is the degree by which the energy of the rain is measured. It is the major factor in soil detachment process. The energy of the rain is

relational to its "erosivity" [1], and it is expressed in Jm−2 mm−1. The technique of varying kinetic energy differs among rainfall simulators and the purpose for which a research is carried out [3]. Obtaining higher kinetic energy with drop forming simulator is an indication of the non-portability of the simulator because it requires higher height get such KE value. Aksoy et al. [33] in an investigation obtained kinetic energy of 21 Jm−2 mm−1, using pressurised nozzle simulator at lower rainfall intensity of 45 mm/h and a height of 2.4 m. By varying the drop diameter from 2.7 to 5.1 mm and height of fall from 0.17 to 2.5 m, similar result was obtained [34].

The kinetic energy of rainfall is depending on two factors; terminal velocity at impact and the spraying nozzle which give intensity. Therefore when a simulator is designed for investigation of potential erosion by simulated rainfall, these aforementioned two factors should be taken note of [29]. This can however, be determined by using Eq. (2) [35].

$$\text{KE} = 0.119 + 0.0873 \log \text{I} \tag{2}$$

where KE is the kinetic energy of the rainfall in (MJha−1 mm−1) and I is the rainfall intensity in (mm/h).

#### **4.5 Rainfall distribution uniformity**

In simulated rainfall on a plot, uniformity is one the most important measure of determining how spatially distributed the rainfall is on a plot to avoid ponding and over saturation on one side [3]. It therefore measures the equal catches of simulation of rainfall [28]. There are factors that sometimes affect uniformity: this includes; wind, slope and altitude [1]. The degree of uniformity is dependent of the rainfall type. It is estimated using the Christiansen uniformity coefficient (*C*u) equation as presented in Eq. (3) [3]

$$C\_u = 1 - \frac{SD}{I\_m} \tag{3}$$

where *C*u is the Christiansen uniformity coefficient; *SD* is the standard deviation of simulated rain over the plot; *I*m is the mean simulated rain intensity.

Eq. (2) can further be expressed as in Eq. (4)

$$CU = 1 - \left[\frac{\sum \, ^\circ X\_i - X\_{m'}}{n X\_m}\right] \ge 100\tag{4}$$

**101**

**5.1 Pressure**

emphasised.

velocity and angle of aperture.

*A Revisit of Rainfall Simulator as a Potential Tool for Hydrological Research*

contrary to sprinkler uniformity standard bench mark of 85% [10].

**5. Design requirements of rainfall simulator**

the simulator from the plot for nozzle that produces cone spray, operating pressure [36] in drop forming simulators (DFs) whereas in pressurised simulator (PN) *C*u is increased based on increase in pressure and intensity [3]. Many researches have been carried out to estimate uniform application of depths as was used by Christensen to investigate the factors affecting water distribution from group of sprinklers [28, 29, 37–39], but this has been recently criticised based on the fact that is less significant and that size of rain gauge for uniformity and intensity affects the results [14], yet it often used as guide in rainfall simulation. Uniformity of can be more than 90% [31]

The methods employed to measure coefficient of uniformity plays a significant

To successfully achieve afore listed natural rainfall characteristics, a designer of a rainfall simulator should take into considerations the following phonotypical features; pump pressure, simulators height, plot size and nozzle spacing. Each these physical features have impact on the purpose for which the rainfall simulator is designed.

In pressurised nozzle simulator the choice of pressure is a determining factor to mimic the natural rainfall to the nearest possible outcome [40]. The basis for selecting pressure should be such that balance is stroked among rain intensity, uniformity, rain drop size and kinetic energy, but different researchers are embedded with different approach toward pressure [3]. For example, in an investigation carried out by Cerda et al. indicated that uniformity was obtained at pressure 152 kPa using HARDI-1553-10 single nozzle and anything above this settings resulted to higher rain concentration at the plot boarder and below resulted to concentration of rain at the centre of the plot. The researcher therefore noted that increase in pressure has a maximum limit when targeting at rain uniformity above which decreases the uniformity [3]. In similar research by Sousa-Junior & Siqueira [31], similar trend of results were observed. Simulator under rainfall intensity of 3.1 mm/min, produced uniformity coefficient of 85% at 40 kPa [36]. Comparing the result of [35] with [41] investigation of rainfall intensity at 20 kPa and achieving 1.42–1.58 mm/min with an average rain uniformity of 60%, therefore, the effect of pressure cannot be over

In Aksoy et al. [33] investigation, the orifice size was appreciated on examining the effects of pressure on 4-Veejet 8030, 4-Veejet 8050, 5-Veejet 8060 and 5-Veejet 8070 nozzles of different orifices, except for 5-Veejet 8060 nozzle which gave rain uniformity of 83.6% at 33 kPa pressure otherwise the others mimicked uniformity of 82.1, 86, and 88.8% at 40, 42 and 48 kPa respectively. Larger orifice resulted to increase in uniformity though with increase in pressure. According to [14], study on development and calibration of pressurised nozzle simulator observed that uniformity and intensity of modelled rainfall are affected by nozzle pressure disc angular

role in achieving correct simulated rainfall data [3]. It is therefore difficult to compare the uniformity results of simulated rainfall from different report [31]. In a review, [5] pointed out that drop forming simulators produces higher rainfall uniformity than pressurised nozzle simulator at lower rain intensity. Generally speaking without considering rainfall simulator type, investigator achieved average rain uniformity of 83% within the intensity range of 10 mm/h and 182 mm/h [1, 3, 31].

*DOI: http://dx.doi.org/10.5772/intechopen.93532*

where *X*i is the individual rain gauge, *X*m is the mean gauge of the rainfall and *n* is total number of rain gauges.

Spray patterns of different types are obtained from different nozzles. In rainfall simulators, there are two different types of nozzles that are often used based on their mould. Namely; flat and cone spray nozzles. From each of these nozzles there tends to be decrease in uniformity from centre to outward of the sprayed plot [3, 24]. The challenge of rainfall uniformity reducing from centre to outward of the plot can be mitigated by using network of nozzles, taking into consideration the wetted perimeter of individual nozzles. The wetted perimeter depends on the distance of

#### *A Revisit of Rainfall Simulator as a Potential Tool for Hydrological Research DOI: http://dx.doi.org/10.5772/intechopen.93532*

the simulator from the plot for nozzle that produces cone spray, operating pressure [36] in drop forming simulators (DFs) whereas in pressurised simulator (PN) *C*u is increased based on increase in pressure and intensity [3]. Many researches have been carried out to estimate uniform application of depths as was used by Christensen to investigate the factors affecting water distribution from group of sprinklers [28, 29, 37–39], but this has been recently criticised based on the fact that is less significant and that size of rain gauge for uniformity and intensity affects the results [14], yet it often used as guide in rainfall simulation. Uniformity of can be more than 90% [31] contrary to sprinkler uniformity standard bench mark of 85% [10].

The methods employed to measure coefficient of uniformity plays a significant role in achieving correct simulated rainfall data [3]. It is therefore difficult to compare the uniformity results of simulated rainfall from different report [31]. In a review, [5] pointed out that drop forming simulators produces higher rainfall uniformity than pressurised nozzle simulator at lower rain intensity. Generally speaking without considering rainfall simulator type, investigator achieved average rain uniformity of 83% within the intensity range of 10 mm/h and 182 mm/h [1, 3, 31].
