Water Science

ScienceDirect
Water Science 31 (2017) 34–43

journal homepage: www.elsevier.com/locate/wsj

Estimation and inter-comparison of infiltration models

Parveen Sihag ∗ , N.K. Tiwari, Subodh Ranjan
Department of Civil Engineering, National Institute of Technology, Kurukshetra, Haryana, India
Received 16 August 2016; received in revised form 7 March 2017; accepted 12 March 2017
Available online 20 April 2017

Abstract
Infiltration models are very helpful in designing and evaluating surface irrigation systems. The main objective of the present work
is estimation and inter-comparison of infiltration models which are used to evaluate the infiltration rates of National Institute of
Technology (NIT)-campus in district of Kurukshetra, Haryana (India) and for this study, field infiltration tests were carried out at
ten different locations comprising of 109 observations by use of double ring infiltrometer. The potential of three infiltration models
(Kostiakov, Modified Kostiakov and US- Soil Conservation Service (SCS)) were evaluated by least–square fitting to observed
infiltration data. Three statistical comparison criteria including maximum absolute error (MAE), Bias and root mean square error
(RMSE) were used to determine the best performing infiltration models. In addition, a novel infiltration model was developed from
field tests data using nonlinear regression modeling which suggests improved performance out of other three models. In case of
nonexistence of observed infiltration data, this novel model can be used to artificially generate infiltration data for NIT campus.
© 2017 National Water Research Center. Production and hosting by Elsevier B.V. This is an open access article under the CC
BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Keywords: Infiltration; Inter-comparison; Double ring infiltrometer; Novel model

1. Introduction

Infiltration is the term applied for the process of water entry into soil (Hillel and Baker, 1988). It is one of the
important components of the hydrological cycle. Water content, field density, suction head, temperature, humidity,
rainfall intensity and type of impurity play an important role in influencing the infiltration rate. Many researchers have
compared the accuracy of the models by comparing the computed and observed infiltration rates (Hopmans, 1995;
Mishra et al., 2003; Chahinian et al., 2005; Turner, 2006; Haghighi et al., 2010). Under specific conditions, a particular
model shows better predictions than others. Machiwal et al. (2006) observed infiltration was well described by the
Philip’s model in wasteland in Kharagpur, India. Návar and Synnott (2000) evaluated the infiltration rates of soils under
four land uses in north-eastern Mexico. Among the infiltration models of Horton, Philip, a modified Kostiakov and
Green-Ampt, the modified Kostiakov model gave the best fit. Mohammed (2006) evaluated the Kostiakov, Horton, and
Philip’s infiltration models under different tillage and rotations in a clay-loam in north-west Iran and reported that the
Horton’s model gave the best prediction of infiltration rate in that area. Under specific conditions, a particular model

∗ Corresponding author. Parveen Sihag (Ph.D Scholar), Department of Civil Engineering, National Institute of Technology, Kurukshetra, Haryana
136119, India.
E-mail addresses: parveen12sihag@gmail.com (P. Sihag), nand nitk@rediffmail.com (N.K. Tiwari), sranjan nitk@yahoo.com (S. Ranjan).

http://dx.doi.org/10.1016/j.wsj.2017.03.001
1110-4929/© 2017 National Water Research Center. Production and hosting by Elsevier B.V. This is an open access article under the CC BY-NC-ND
license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
 P. Sihag et al. / Water Science 31 (2017) 34–43 35

shows better predictions than others. But till date, it is not specific mentioned which model gives better prediction
(Turner, 2006). Thus, the main aim of present study is to determine the best model for the soil of NIT-campus in district
of Kurukshetra, Haryana (India).

1.1. Infiltration models and parameters

Infiltration models can generally be classified into three groups (Mishra et al., 2003) as below:

1. Physical models: These are deduced from the law of mass conservation and the Darcy’s law (e.g. Green and Ampt,
1911; Richards, 1931; Philips, 1957).
2. Semi-empirical models: These consist of simple hypotheses about the infiltration rate and cumulative infiltration
relation (e.g. Holtan, 1961; Singh and Yu, 1990).
3. Empirical models: These are based on field data and laboratory experiments (e.g. Kostiakov, 1932; Horton, 1941).

1.1.1. Kostiakov model

f (t) = at −b (1)
where f(t) is the infiltration rate (LT−1 ) as a function of time, a and b are the equation’s parameters and t is time (T).

1.1.2. Kostiakov modified model

The modified model of Kostiakov for long time is defined as:
f (t) = at −b + ic (2)
where f(t) is the infiltration rate (LT−1 ) as a function of time, a and b are the equation’s parameters and i c is the steady
infiltration rate (LT−1 ), t is time (T).

1.1.3. SCS model

Soil Conservation Service (SCS) (Jury et al., 1991), is expressed as follows:
I = at b + 0.6985 (3)
where a and b are the equation constants. I is the cumulative infiltration rate (L).

1.1.4. Novel model

A typical non linear regression model has been adopted which is developed from Kostiakov modified model as
shown below:
f (t) = at −b + c ic (4)
where f(t) Soil infiltration rate (LT−1 ) at time t (T), ic steady infiltration rate (LT−1 ). a, b and c are the equation
constants.

1.2. Estimation and inter-comparison of model parameters

For estimation and inter-comparison of models, it is necessary to define the criteria by which inter-comparison is
evaluated. To judge the estimated accuracy, the fallowing statistical parameters are used for quantifying the errors.

1.2.1. Maximum absolute error (MAE)

The maximum absolute error is used to measure of success of numeric estimation. The maximum absolute error
(MAE) is computed as
MAE = max(|x − y| (5)
where
x = observed data values.
36 P. Sihag et al. / Water Science 31 (2017) 34–43

y = Estimated (computed) data values.

1.2.2. Bias
The bias of an infiltration model is the average difference between the predicted value and the observed value of
the infiltration models. The value of zero bias is called unbiased. Its value is defined by:
ni (x − y)
Bias = (6)
n

1.2.3. Root mean square error (RMSE)

Mean-squared error is the most commonly used measure of success of numeric estimation, and root mean-squared
error is the square root of mean-squared-error after we give it the same dimensions as the estimated values themselves.
This method exaggerates the estimated error—the difference between estimated value and observed value (actual
value). The root mean squared error (RMSE) is computed as:
  
 n
1 
RMSE =  (x − y)2 (7)
n
i=1

2. Study area

NIT campus is spread up in 300 acres and it is situated in district of Kurukshetra, which lies about 160 km north
of Delhi, and 40 km south of Ambala. Its adjoining districts are Ambala, Yamuna Nagar, Karnal, Kaithal and Patiala
(Punjab). It is located at 29.9655◦ N, 76.7106◦ E in the central part of Haryana (India). The climate of Kurukshetra
district is mainly dry with very hot in summer and cold in winter, except during monsoon season when moist air of
oceanic origin penetrates into the area. There are four seasons in a year. The normal annual rainfall of the district
is 582 mm which is unevenly distributed over the year. The south west monsoon sets in from last week of June and
withdraws in end of September, contributed about 81% of annual rainfall. July and August are the wettest months. Rest
19% rainfall is received during non-monsoon period in the wake of western disturbances and thunder storms. The study

Fig. 1. Location map of the study area.

 P. Sihag et al. / Water Science 31 (2017) 34–43 37

area falls in Upper-Ghaggar Basin. The entire district of Kurukshetra is covered by tropical arid brown soils. These
soils are pale brown in color. These soils are medium to high in organic matter. Generally, three types of soil, normally
sandy loam, loam and clay loam. The infiltration rate of these soils is low to moderate. The available moisture holding
capacity of these soils is medium to high. The ten different locations were selected for experimentation in National
Institute of Technology, Kurukshetra campus (Fig. 1) to study the variation of infiltration rate.

3. Methodology

Double ring infiltrometer was used for measurement of infiltration rates at all selected locations. As depicted in
Fig. 1, the double ring infiltrometer has two concentric rings of depth 30 cm each, and diameter of 30 cm and 45 cm
for inner and outer rings respectively. The rings were driven at about 10 cm deep into soil by using falling weight
type hammer striking uniformly on a wooden plank placed on top of ring without undue disturbance to soil surface.
The rings were filled with water to the equal depth and the initial reading of water level was noted. The depth of
water in the infiltrometer was noted at regular intervals until the rate of infiltration became steady. The experimental
observation of initial and final infiltration rate and moisture content had been listed in Table 1 for all tests. The soil
sample for calculating moisture content (about 100–150 gm) was collected from a site closest to the location selected
for experimentation.

Site: Abhimanyu Bhawan (Test No - 1) Site: Fanibhushan Bhawan (Test No - 2)

60 Observed Infiltraon rate Kosakov model 90
Observed Infiltraon rate Kosakov model
50 Kosakov modified model SCS model 80
70 Kosakov modified model SCS model
40 Novel model 60
Infiltraon Rate (mm/hr)

Novel model
Infiltraon Rate (mm/hr)

30 50
40
20 30
10 20
10
0
0
0 50 100 150 200
0 50 100 150 200

Elapsed Time (minutes) Elapsed Time (minutes)

Site: Bhishma Bhawan (Test No - 3) Site: Ekalavya Bhawan (Test No - 4)

70 120 Observed Infiltraon rate Kosakov model
60 Observed Infiltraon rate Kosakov model
100 Kosakov modified model SCS model
Infiltraon Rate (mm/hr)

Kosakov modified model SCS model

50
Novel model 80 Novel model
40
Infiltraon Rate (mm/hr)

60
30
40
20
10 20
0 0
0 50 100 150 200 0 50 100 150 200

Elapsed Time (minutes) Elapesd Time (minutes)

Site: Dronacharya Bhawan (Test No - 5)

140 Observed Infiltraon rate Kosakov model
120 Kosakov modified model SCS model
100 Novel model

80
Infiltraon Rate (mm/hr)

60
40
20
0
0 50 100 150 200

Elapse Time (minutes)

Fig. 2. Inter-comparison of field infiltration rates and various models estimated infiltration rate for study area.
38 P. Sihag et al. / Water Science 31 (2017) 34–43

Site: Girivar Bhawan (Test No - 6) Site: Harihar Bhawan (Test No - 7)

60 70 Observed Infiltraon rate Kosakov model
Observed Infiltraon rate Kosakov model
50 Kosakov modified model SCS model 60 Kosakov modified model SCS model
Novel model 50
40 Novel model

Infiltraon Rate (mm/hr)

40
Infiltraon Rate (mm/hr)

30
30
20 20
10 10
0
0
0 50 100 150 200
0 50 100 150 200

Elapse Time (minutes) Elapsed Time (minutes)

Site: Indivar Bhawan (Test No - 8) Site: Chakaradhar Bhawan Test No - 9)

80 70 Observed Infiltraon rate Kosakov model
Observed Infiltraon rate Kosakov model
70 60 Kosakov modified model SCS model
Kosakov modified model SCS model
60

Infiltraon Rate (mm/hr)

50 Novel model
Novel model
Infiltraon Rate (mm/hr)

50
40
40
30
30
20 20
10 10
0 0
0 50 100 150 200 0 50 100 150 200

Elapsed Time (minutes) Elapsed Time (minutes)

Site: Civil Engineering Department (Test No - 10)

70 Observed Infiltraon rate Kosakov model
60 Kosakov modified model SCS model
50
Infiltraon Rate (mm/hr)

Novel model
40
30
20
10
0
0 50 100 150 200

Elapsed Time (minutes)

Fig. 2. (Continued)

Table 1
Details of initial, final infiltration rates and moisture contents of study area.
Test no. Location Initial infiltration rate (mm/hr) Final infiltration rate (mm/hr) Moisture content (%)

1 Abhimanyu Bhawan 48 15 3.44

2 Fanibhushan Bhawan 60 15 2.65
3 Bhishma Bhawan 48 10 1.93
4 Ekalavya Bhawan 96 5 7.98
5 Dronacharya Bhawan 84 4 7.65
6 Girivar Bhawan 48 9 3.51
7 Harihar Bhawan 60 8 3.40
8 Indivar Bhawan 60 8 4.20
9 Chakaradhar Bhawan 48 11 5.27
10 Civil Engineering Department 48 11 3.83
 P. Sihag et al. / Water Science 31 (2017) 34–43 39

4. Results

Each infiltration test was carried out in the field in order to deal with the spatial variability of infiltration rate.
Based on the field tests at 10-location in district of Kurukshetra at NIT-campus, results were analysed and individual
infiltration curves have been generated. Results of field infiltration rates and estimated (computed) infiltration rates
were summarized in Fig. 2. The final infiltration rates for 10 locations are 15, 15, 10, 5, 4, 9, 8, 8, 11 and 11 mm/hr
respectively. Table 1 indicates the values of initial infiltration rate, final infiltration rates and moisture contents of the
soil samples of different locations. The initial infiltration rate varies from 96 mm/hr to 48 mm/hr and moisture contents
of the soil samples taken from the sites were varies from 7.98% to 1.93%. It was found that the location at which
moisture content was higher it took less time to achieve constant infiltration rates.
Water infiltration data for different soils are essential for understanding of the rainfall-runoff process. A number of
algebraic models have been proposed to determine field infiltration rates. The proposed models Kostiakov, Kostiakov
modified, SCS and the novel model were chosen for evaluation in the study. To study these models, observed field
infiltration data have been used. Attempts were made to evaluate these infiltration models on the basis of experimental
data of the study area and to obtain numerical values for the parameters of the models. XLSTAT software has been
used for deducing the parameters of the infiltration models and analysing infiltration data using least square technique.
The details of infiltration models parameters were summarized in Table 2 and graphs of estimated infiltration rate and
relative error are presented in Figs. 2 and 3.

Site: Abhimanyu Bhawan (Test No - 1) Site: Fanibhushan Bhawan (Test No - 2)

10 Kosakov model Kosakov modified model 20 Kosakov model Kosakov modified model
SCS model Novel model
SCS model Novel model 15
5 10
5
0 0
Relave Error
Relave Error

0 50 100 150 200 -5 0 50 100 150 200

-5 -10
-15
-10 -20

Elapsed Time (minutes) Elapsed Time (minutes)

Site: Bhishma Bhawan (Test No - 3) Site: Ekalavya Bhawan (Test No - 4)

15 Kosakov model Kosakov modified model 15 Kosakov model Kosakov modified model
SCS model Novel model SCS model Novel model
10 10

5 5
Relave Error

0
Relave Error

0
0 50 100 150 200 0 50 100 150 200
-5 -5
-10 -10
-15 -15
Elapsed Time (minutes) Elapsed Time (minutes)

Site: Dronacharya Bhawan (Test No - 5)

40 Kosakov model Kosakov modified model
30 SCS model Novel model
20
10
Relave Error

0
-10 0 50 100 150 200

-20
-30
-40

Elapsed Time (minutes)

Fig. 3. Distribution of relative errors of different infiltration models for study area.
40 P. Sihag et al. / Water Science 31 (2017) 34–43

Site: Girivar Bhawan (Test No - 6) Site: Harihar Bhawan (Test No - 7)

10 Kosakov model Kosakov modified model 10 Kosakov model Kosakov modified model
SCS model Novel model 8 SCS model Novel model
5 6
4
2

Relave Error
0
Relave Error

0
0 50 100 150 200
-2 0 50 100 150 200
-5 -4
-6
-10 -8

Elapsed Time (minutes) Elapsed Time (minutes)

Site: Indivar Bhawan (Test No - 8) Site: Chakaradhar Bhawan (Test No - 9)

15 Kosakov model Kosakov modified model 15 Kosakov model Kosakov modified model
SCS model Novel model SCS model Novel model
10 10

5 5

0
Relave Error

Relave Error
0 50 100 150 200 0 50 100 150 200
-5 -5

-10
-10
-15
-15
Elapsed Time (minutes) Elapsed Time (minutes)

Site: Civil Engineering Department (Test No - 10)

15 Kosakov model Kosakov modified model
10 SCS model Novel model

0
Relave Error

0 50 100 150 200

-5

-10

-15
Elapsed Time (minutes)

Fig. 3. (Continued)

Table 2
Parameters of the selected infiltration models.
Test no. Parameters of the infiltration models

Kostiakov model Kostiakov modified model SCS model Novel model

a b a b a b a b c

1. 80.42 0.36 131.85 0.86 1.13 0.75 121.59 0.79 0.93

2. 190.47 0.35 123.13 0.58 1.95 0.69 115.72 0.23 −1.27
3. 104.96 0.46 134.87 0.75 1.55 0.65 103.88 0.46 −0.03
4. 638.90 1.19 921.63 1.44 4.62 0.36 832.44 1.37 0.74
5. 349.56 0.86 397.95 0.96 5.39 0.36 294.04 0.71 −1.69
6. 108.57 0.56 206.41 1.04 1.23 0.65 194.00 1.00 0.94
7. 178.87 0.70 290.19 1.06 1.90 0.58 264.78 0.99 0.85
8. 158.03 0.62 225.91 0.91 2.14 0.57 196.71 0.81 0.71
9. 95.58 0.44 135.09 0.79 1.40 0.68 108.55 0.59 0.57
10. 95.04 0.44 135.09 0.79 1.40 0.68 108.55 0.59 0.57

a, b and c: Equation parameters.

 P. Sihag et al. / Water Science 31 (2017) 34–43 41

Table 3
Inter—comparison parameters of infiltration models.
Sr.no. Test no. Kostiakov model Kostiakov modified model SCS model Novel model

(i) Maximum absolute error (MAE)

1 1 6.279 3.975 5.474 4.249
2 2 6.220 7.905 19.915 5.417
3 3 6.453 8.112 13.357 6.439
4 4 5.796 6.673 10.616 6.550
5 5 11.184 12.433 39.559 9.432
6 6 5.969 3.590 3.884 3.831
7 7 4.181 2.716 7.875 2.375
8 8 5.282 3.069 12.099 3.806
9 9 4.941 3.753 10.479 4.028
10 10 4.911 3.753 10.479 4.028
Average 6.122 5.598 13.374 5.016
(ii) Bias
11 1 −0.191 0.081 −0.209 0.0000857000
12 2 0.061 0.387 −0.086 0.0005670000
13 3 −0.004 0.484 −0.203 −0.0000490000
14 4 −0.853 0.391 −0.331 0.0000012800
15 5 0.626 1.271 −0.443 0.0002910000
16 6 −0.293 0.079 −0.058 0.0000734000
17 7 −0.464 0.192 −0.058 0.0000004650
18 8 −0.232 0.229 −0.002 0.0000000588
19 9 −0.072 0.277 0.105 −0.0000048000
20 10 −0.081 0.277 0.105 −0.0000048000
Average −0.1503 0.3668 −0.118 0.0000960300
(iii) Root mean square error (RMSE)
21 1 2.771 2.022 2.855 2.003
22 2 2.432 3.046 5.527 2.431
23 3 3.079 2.562 5.250 2.464
24 4 4.164 4.976 15.22 3.692
25 5 2.706 1.778 1.715 1.766
26 6 2.447 1.427 3.160 1.349
27 7 2.109 1.836 4.309 1.696
28 8 1.944 2.043 3.893 1.765
29 9 1.944 2.043 3.893 1.765
30 10 1.944 2.043 3.893 1.765
Average 2.554 2.378 4.971 2.070

Table 4
Result of ANOVA single factor test.
Infiltration models F-value P-value F critical

All models 0.021835 0.995621 2.62555

Kostiakov and novel model 0.0046716 0.945570 3.88487

Infiltration models were evaluated using Maximum absolute error (MAE), Bias and Root mean square error (RMSE)
statistical criteria. The best fit model was selected on the basis of minimum the Maximum absolute error (MAE), Bias
and Root mean square error (RMSE) criteria. Findings are summarized in Table 3.
The estimated average values of Maximum absolute error (MAE) were 6.122, 5.598, 13.374, 5.016, Bias were
−0.1503, 0.3668, −0.118, 0.000096, and root mean square error (RMSE) values were 2.554, 2.378, 4.971, 2.070
for Kostiakov, Kostiakov modified, SCS and a novel model respectively. Results from Table 4 suggests that F-value
(0.021835) was less than f-critical (2.62555) and P-value (0.995621) was greater than 0.05 suggesting that difference
in predicted values was insignificant for all the models.
42 P. Sihag et al. / Water Science 31 (2017) 34–43

140
Kosakov model
120
Kosakov modified model

Predicted Infiltraon Rate (mm/hr)

100 SCS model

80 Novel model

60

40

20

0
0 20 40 60 80 100 120 140

Observed Infiltraon Rate (mm/hr)

Fig. 4. Observed vs. estimated infiltration rates using various models.

Fig. 4 shows a graph between observed and estimated infiltration rates using above the models. To study the scatter
around the line of perfect agreement, two more lines in the range of ±20% errors between the observed and estimated
(predicted) values of infiltration rate were plotted in the resulting graph. Fig. 4 indicates that the most of the values
predicted by a novel model lie within ±20% error bands from the line of perfect agreement.
Results from Table 4 suggests that F-value (0.0046716) was less than f-critical (3.88487) and P-value (0.945570)
was greater than 0.05 suggesting that difference in predicted values was insignificant for Kostiakov and novel model.
Comparison of the statistical parameters MAE, Bias and RMSE suggests that the novel model is performing better than
other models and it may be used to assess the infiltration rate with similar field characteristics of N.I.T Kurukshetra,
Haryana, India.

5. Conclusion

Infiltration models have become one of the important tools for the management of water resources systems. This
paper investigates the ability of the novel infiltration model in estimating the infiltration rate from actual field data in
comparison to other three popular models in like situations. The results presented in this work are quite encouraging.
Further the findings of paper show that the novel model is the most suitable amongst other three models for estimation
of infiltration rate in specific land use and soil parameters. From Table 3 the average value of MAE, Bias and RMSE of
novel infiltration model were 5.016, 0.000096 and 2.070 respectively achieved. The novel infiltration model average
value of MAE, Bias and RMSE were lower among all other infiltration models. From comparison of actual field data
and data estimated by the novel infiltration model, it may be concluded that it is closely fitted amongst other three
models and general nature of both data sets is also more or less similar. The novel infiltration model is as simple as
modified Kostiakov model and it is more accurate than other infiltration models for the study area. Efforts should be
made out to gather infiltration data for longer periods so that the model can be further refined and predictions made
with more accuracy. ’

References

Green, W.H., Ampt, G., 1911. Studies on soil physics,1. The flow of air and water through soils. J. Agr. Sci. 4, 1–24.
Chahinian, N., Moussa, R., Andrieux, P., Voltz, M., 2005. Comparison of infiltration models to simulate flood events at the field scale. J. Hydrol.
306 (1), 191–214, http://dx.doi.org/10.1016/j.jhydrol.2004.09.009.
Haghighi, F., Gorji, M., Shorafa, M., Sarmadian, F., Mohammadi, M.H., 2010. Evaluation of some infiltration models and hydraulic parameters.
Span. J. Agric. Res. 8 (1), 210–217, http://dx.doi.org/10.5424/sjar/2010081-1160.
Hillel, D., Baker, R.S., 1988. A descriptive theory of fingering during infiltration into layered soils. Soil Sci. 146 (1), 51–56, 0038-075X/88/1461-0051.
Holtan, H.N., 1961. Concept for Infiltration Estimates in Watershed Engineering, ARS41-51. U.S. Department of Agricultural Service, Washington,
DC.
 P. Sihag et al. / Water Science 31 (2017) 34–43 43

Hopmans, J., 1995. Evaluation of various infiltration models. Sci. Agric. 140, 5–8.
Horton, R.E., 1941. An approach toward a physical interpretation of infiltration-capacity. Soil Sci. Soc. Am. J. 5 (C), 399–417.
Jury, W.A., Gardner, W.R., Gardner, W.H., 1991. Soil Physics, fifth ed. John Wiley & Sons, New York (NY).
Kostiakov, A.N., 1932. On the dynamics of the coefficient of water-percolation in soils and on the necessity for studying it from a dynamic point of
view for purposes of amelioration. Trans 6, 17–21.
Machiwal, D., Jha, M.K., Mal, B.C., 2006. Modelling infiltration and quantifying spatial soil variability in a wasteland of Kharagpur, India. Biosyst.
Eng. 95 (4), 569–582, http://dx.doi.org/10.1016/j.biosystemseng.2006.08.007.
Mishra, S.K., Tyagi, J.V., Singh, V.P., 2003. Comparison of infiltration models. Hydrol. Processes 17 (13), 2629–2652,
http://dx.doi.org/10.1002/hyp.1257.
Mohammed, A.H., 2006. Evaluation of Kostiakov, Horton and Philip’s infiltration equations as affected by tillage and rotation systems in a clay-loam
soil of Northwest Iran. In: The 18th World Congress of Soil Science, July.
Návar, J., Synnott, T.J., 2000. Soil infiltration and land use in Linares, NL, Mexico. Terra 18 (3), 255–262.
Philips, J.R., 1957. The theory of infiltration: the infiltration equation and its solution. Soil Sci. 83 (5), 345–357.
Richards, L.A., 1931. Capillary conduction of liquids through porous mediums. J. Appl. Phys. 1 (5), 318–333.
Singh, V.P., Yu, F.X., 1990. Derivation of infiltration equation using systems approach. J. Irrig. Drain. Eng. 116 (6), 837–858,
http://dx.doi.org/10.1061/(ASCE)0733-9437(1990)116:6(837).
Turner, E.R., 2006. Comparison of Infiltration Equations and Their Field Validation with Rainfall Simulationcomparison of Infiltration Equations
and Their Field Validation with Rainfall Simulation (Doctoral dissertation).