The experiment was conducted at the Agricultural Research and Experimental Farm in Dirab, Riyadh, Saudi Arabia (lat. 24.4195° N, long. 46.65° E, and 552 m above sea level elevation) from June to September 2018. Monthly averages of climatic data during experimental period are described in Table 1. The average air temperatures recorded were between 34.6 and 37.4°C, whereas the means of relative humidity recorded were between 10.1% and 13.7%. The recorded intermediate maximum wind speeds fall approximately between 6.6 and 5.6 m s-¹, and the recorded mean solar radiations fall between 23.2 and 24.8MJ m^-2 day^-1. Finally, there was no rainfall during the experimental months.

To investigate the physical and chemical properties of the soil, three samples were collected from different depths in various plots. Table 2 presents the values of the soil texture, field capacity (FC), wilting point, soil bulk density, and initial water content at different soil depth levels from the experimental locations. Finally, the chemical properties of the soil samples from different experimental sites are given in Table 3. The chemical properties of the irrigation water were analyzed by knowing an electrical conductivity value of 2.5 mS cm^-1, pH of 7.48, and total dissolved solids of 2880 mg L^-1. Both the soil and water present in the experimental samples were of reasonable quality to conduct the present study.

The field experiments were designed and executed as follows: the irrigation system was fixed by incorporating a tanks, pump unit, pressure gauges, ball valve, filtration system, pressure regulator, air relief valve, control panel, flow-meter, solenoid valve, main lines, sub-main lines, drip lines, connectors, and line end-caps (Fig. 1). This system was made using a main PVC pipe with an inside diameter of 75 mm that had direct contact with the main water source. A second PVC pipe with an inner diameter of 21 mm was used to transfer the water to the drip lines. The drip lines had an inner diameter of 16 mm and a thickness of 1 mm. The drip lines were buried manually in SDI. Two fiberglass tanks were used to pour water into the network. The first tank had a capacity of 2000 L and the second tank had a capacity of 5000 L. Two drip line depths from the soil surface (D_L) of 15 and 25 cm and DI were applied in three blocks. Each block was divided into three plots to randomly allocate the three soil surface mulching treatments (NM = no mulching, BPF mulching, and PTW mulching), in a randomized complete block design (RCBD) with three replications. Each plot had three experimental units. Each one had three drip lines of 4.5 m long, 70 cm spacing. The distance between emitters was 30 cm. The in-line emitter discharge was 4 L h¹ at operating pressure of 150 kPa.

A water irrigation schedule was created to supply a reasonable amount of water via proper management. The basic aim was to intensify the effectiveness of the irrigation by providing enough water that could increase the θ_v to an adequate level.

The irrigation time was changeable owing to the planned irrigation treatment. The sensors were used to monitor the θ_v before and after irrigation. Scheduling consisted of applying the right amount of water at the right time. Its purpose was to maximize the irrigation efficiency by applying the appropriate amount of water needed to replenish the soil water to the desired level. In the present study, because there were no crops planted, the applied water was controlled based on the FC of the soil. The water depth was calculated for each soil depth from 10 to 50 cm and cumulated. The depth of water added to reach the soil FC (D_w) was calculated using Eq. (1):

For constant monitoring of the water content in the soil, EasyAG probes (Sentek Sensor Technologies, Stepney, Australia) were installed, which provide a θ_v profile for irrigation and management applications. These probes include several sensors that measure the soil water at multiple depths. The probes create a high-frequency electrical field around each sensor that extends through the assessment tube into the soil. The electrical capacitance from the probe provided a θ_v. This was converted from a scaled frequency reading (Eq. 2) using a calibration equation (Eq. 3), which was based on field data:

An analysis of variance following a RCBD was conducted on the average θ_v using the SAS statistical package to determine the effects of treatment (D_L and mulching type, M) on the measured parameters. The treatment means were separated through a least significant difference (LSD) test with a level of statistical significance of 0.05.

Fig. 3 shows that 89.45% and 94.04% of water in NM treatment were applied in the BPF and PTW treatments, respectively, at the DI (D_L = 0 cm). The BPF and PTW treatments at D_L of 15 cm are 6.79% and 4.94% water savings, respectively, whereas approximately 69 mm of water was applied in the NM treatment. At D_L of 25 cm, the quantity of water applied in the NM treatment was ~ 73 mm; 7.02% and 5.26% water savings were achieved in the BPF and PTW treatments, respectively. As shown in Fig. 3, under any type of mulching, the amount of water was higher when the DI was used (i.e., D_L = 0 cm) because of higher evaporation rates from the soil surface (Al-Ghobari & El-Marazky, 2012; Çolak et al., 2018). SDI (i.e., D_L of 15 cm and 25 cm) under any type of mulching was saved along with the applied water. The amounts of applied water at D_L of 15 cm were 5.26%, 5.03%, 4.94%, respectively, lower than those at D_L of 25 cm for NM, BPF, and PTW mulching. Therefore, a BPF or PTW mulching combined with SDI retains the moisture and decreases the required water amount to prevent water evaporation from the soil surface (Gan et al., 2013). However, it is better to use PTW mulch, because it does not require any additional costs, at D_L of 15 cm.

Figure 4 shows the average θ_v values in the soil depths for DI, D_L of 15 cm and D_L of 25 cm under NM, BPF, and PTW treatments. The BPF treatment had higher θ_v values than that of the NM and PTW treatments in both the DI and SDI systems. For the DI with BPF mulching, the θ_v values were approximately 14.79%, 13.69%, and 13.27% directly at the emitter (S of 0 cm), S of 15 cm, and S of 30 cm, respectively (Fig. 4). The θ_v values for the PTW treatment were 14.55%, 13.63%, and 13.15%, at S of 0 cm, S of 15 cm, and S of 30 cm, respectively. The θ_v values for the BPF treatment were 4.08%, 1.33%, and 1.76% higher than that of the NM treatment at S of 0 cm, S of 15 cm, and S of 30 cm, respectively. The θ_v values for the PTW treatment were 2.39%, 0.89%, and 0.84% higher than that of the NM for S of 0 cm, S of 15 cm, and S of 30 cm, respectively.

For D_L of 15 cm, the θ_v values for the BPF treatment were also higher than those of the NM and PTW treatments (Fig. 4). The θ_v values at S of 0 cm were approximately 15.05% and 14.72% for the BPF and PTW treatments, respectively (i.e., the θ_v values were 3.65% and 1.38% higher, respectively, than that of the NM treatment). The θ_v values for the BPF and PTW treatments were 1.02% and 0.58% higher, respectively, at S of 15 cm than that of the NM treatment, whereas the θ_v values increased by 1.58% and 0.75%, respectively, at S of 30 cm. For D_L of 25 cm, the θ_v values at S of 0 cm were 2.74% and 1.13% higher for the BPF and PTW treatments, respectively, than that of the NM treatment (Fig. 4). Additionally, the θ_v values increased by 3.97% and 2.91%, respectively, at S of 15 cm, and the θ_v values increased by 3.15% and 2.20%, respectively, at S of 30 cm.

The comparison of θ_v values average across the experimental treatments is summarized in Table 5. The M had very significant (p < 0.01) effects on the average θ_v values when measured at S of 0 cm and S of 15 cm and significant (p < 0.05) effect at S of 30 cm, irrespective of the D_L treatments. The BPF treatment provided a higher average θ_v value than that of the NM treatment, with a significant increase of 3.50%, 2.18%, and 2.25% at S of 0 cm, S of 15 cm and S of 30 cm, respectively. This was because that BPF mulching stopped the movement of water vapor from the soil surface to the air (Dong et al., 2018). When the PTW covered the soil, the increase in θ_v was significant (1.65% and 1.45%, respectively) at S of 0 cm and S of 15 cm and insignificant at S of 30 cm, comparing with NM treatment, consistent with the results of Liu et al. (2018). This was because the rate of water vapor flux through covered PTW was slow compared to the rate of water loss from wet soil surface (Li R et al., 2013; Li S et al., 2013). BPF treatment showed significant increases (1.82%) in θ_v compared to PTW at S of 0 cm, but significant difference were not observed at S of 15 cm and S of 30 cm. Thus M, which had an effect in the covered soil, retained higher moisture levels, leading to better root growth than that of the uncovered soil. Although it is cost-effective to purchase BPF, this type of mulching system can be replaced by PTW, which is available to farms at no extra cost.

Figure 4 shows that the D_L of 25 cm for the NM treatment had the highest average θ_v value (14.98%) at S of 0 cm, which was 5.42% and 3.17% higher than that at DI and D_L of 15 cm, respectively. The θ_v values at S of 15 cm for the NM treatment were approximately 13.79% and 14.09% for D_L of 15 cm and 25 cm, respectively, i.e., the θ_v values were 2.07% and 4.29% higher, respectively, than that of the DI system. The θ_v values at S of 30 cm for DI were 2.03% and 4.54% lower than that of the D_L of 15 cm and 25 cm, respectively. This result is consistent with Mokh et al. (2014), who explained that the D_L in SDI system influenced θ_v values during the two cropping periods of potato, and increasing the D_L lead to increased θ_v values.

For BPF and PTW treatments, Fig. 4 shows that the θ_v values for the DI system were lower than those of the SDI system. A D_L of 25 cm with the BPF treatment produced the highest θ_v value of 15.39% at S of 0 cm compared to the DI and D_L of 15 cm, which was 4.06% and 2.26% higher, respectively, whereas at S of 15 cm θ_v values increased by 7.01% and 5.17%, and at S of 30 cm values increased by 6.18% and 4.22%. The θ_v values in the PTW treatment under different D_L showed a similar trend, being 4.12%, 6.38%, and 6.16% higher for D_L of 25 cm than those of DI at S of 0 cm, S of 15 cm, and S of 30 cm, respectively. The θ_v values for D_L of 15 cm were 2.84%, 4.34%, and 3.94% lower than those of D_L of 25 cm at S of 0 cm, S of 15 cm, and S of 30 cm, respectively.

Irrespective of M treatments, Table 5 shows that D_L had a significant (p < 0.01) effects on the average θ_v values at different S, being D_L of 25 cm the treatment showing the highest value, unlike in the DI. Significant differences between D_L treatments were observed at S of 0 cm, S of 15 cm and S of 30 cm, the average θ_v value at D_L of 25 cm were 4.27%, 5.95%, and 5.78% higher than those of the DI, while 2.57%, 4.04%, and 3.73% higher than those of the D_L of 15 cm, respectively. The θ_v values’ variance between D_L of 15 cm and D_L of 25 cm treatments are only small. So, the D_L should be at 15 cm to reduce the cost of drilling. The deepening of the drip line away from the sun results in increasing θ_v value due to a lack of moisture loss (Solomon, 1993).

Figure 5 show that the θ_v distribution was affected by M and D_L under different S. The best uniformity of θ_v distribution contour lines throughout the soil profile was obtained under SDI (D_L of 15 cm and D_L of 25 cm). However, the distribution of the θ_v for different M treatments indicated that the D_L of 15 cm and D_L of 25 cm had more uniform bulb distribution at S of 0 cm. In contrast, the θ_v distribution in S of 15 cm and S of 30 cm was similar and more uniform than that obtained with the DI. The θ_v bulb’s spread decrease as S increases horizontally under any M and any D_L. Similarly, Assouline (2002), Grabow et al. (2006), Badr (2007), Shirahatti et al. (2007), and Nasrabad et al. (2013) showed that the θ_v value decreased horizontally as the S increased. In sandy soil, the emitters need to be closer together because the water does not move as far horizontally (Arbat et al., 2010). Moreover, in an SDI system, the vertical movement of the θ_v level was found to be higher than the horizontal movement (Bajracharya & Sharma, 2005; Al-Ghobari & El-Marazky, 2012; Douh et al., 2013).

Table 5 shows that binary interactions between the M and D_L had (p < 0.05) significant effect on the θ_v values at S of 15 cm only. BPF and PTW mulching at 0-20 cm soil layer increased θ_v values by 0.96% and 1.25%, respectively, more than that of NM in DI (Fig. 5). The corresponding values of θ_v were increased by 3.05% and 2.91% for D_L of 15 cm while 5.56% and 5.41% for D_L of 25 cm. This agrees with Wang et al. (2009) and Liu et al. (2014). Ma et al. (2018) found that plastic film mulching increased the θ_v significantly (12.9%) for the 0-20 cm soil layer, compared with traditional approach. Using mulching (e.g., BPF and PTW) holds water evaporation and encourages water movement to the topsoil layers promoting θ_v during initial stage of crop growth (Gan et al., 2013). With SDI, the surface soil layer is not completely wetted (i.e. lower moisture) as in the case of DI. Therefore, with SDI the upper soil layers remain relatively dry, thereby reducing the direct soil evaporation as compared to DI (Solomon, 1993). At the 0-40 cm soil layer, being the normal root depth for most crops, the average θ_v value in S of 15 cm was 13.87% for the D_L of 15 cm and 14.13% for the D_L of 25 cm under NM treatment (Fig. 5a). This observation agrees with the results reported by Badr & Abuarab (2011), who suggested that a D_L of 30 cm is deemed the active root zone in vegetable crops, and the improved activity was attributed to the enhanced capacity to restore water, particularly for sandy soils. In contrast, a D_L of greater than 10 cm is advisable to prevent the wetting of the soil surface during irrigation in loamy soil (Rodríguez-Sinobas et al., 2012). The corresponding values were 14.02% and 14.63% under BPF treatment (Fig. 5b), while 13.97% and 14.54% under PTW treatment (Fig. 5c). The increased moisture retention capacity of BPF and PTW treatments could be attributed to less non-productive water losses from the soil, which play a vital role in the management and growth of crop (Zhao et al., 2014; Dong et al., 2018; Li et al., 2018). Because of vapors, the water was further trapped within the mulch, resulting in fog, which again dropped into the upper soil layer, as reported by Ashrafuzzaman et al. (2011). The θ_v distribution contours show a saturation bulb under the emitters that moves downward as the D_L increases (Fig. 5). Clearly, the θ_v distribution became more controllable moving downward when applying BPF and PTW at D_L of 25 cm than at D_L of 15 cm. This is consistent with Thorburn et al. (2003), who showed that the D_L controlled the amount of water reaching the surface and the upward spread of the θ_v toward the soil surface. The shape of the bulb also changed from a near-circle to an ellipse when the BPF and PTW mulching were used. BPF mulching at D_L of 25 cm largely allowed the downward movement of θ_v (Fig. 5b). Fig. 5c shows similar results but with less θ_v moved downward when the PTW mulching was applied at D_L of 25 cm. It is better to use D_L at 15 cm and PTW mulching, that is less expensive to install, giving slightly less θ_v values than those of BPF mulching at D_L of 25 cm.

In summary, the present study illustrated the influence of the D_L under different M in a SDI system for the θ_v distribution in a soil profile. The inclusion of BPF or PTW mulching on the soil surface was found to enhance the water retention capacity of the soil. The SDI system reduced the required water amount when the drip line was mulched with BPF by a small value compared to when the drip line was mulched with PTW. Therefore, it is recommended that the methodology of an SDI system would provide a useful method for treating soil through the installation of a D_L at 15 cm and by mulching the soil with PTW where no additional cost is required. Such treatment will provide an active zone of soil to the roots of vegetables crops. Therefore, we believe that the soil treatment strategy outlined in the present study could restore high levels of water resources in the loamy land of Saudi Arabian farms at a significantly low cost.