Assessing the Role of Rainfall Redirection Techniques for Arresting the Land T Degradation Under Drip Irrigated Grapevines ⁎ V

Journal of Hydrology 587 (2020) 125000

Contents lists available at ScienceDirect

Journal of Hydrology

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

Research papers Assessing the role of rainfall redirection techniques for arresting the land T degradation under drip irrigated grapevines ⁎ V. Phogata,b,c, , T. Pitta,b, R.M. Stevensd, J.W. Coxa,b, J. Šimůneke, P.R. Petriea,b,f a South Australian Research and Development Institute, GPO Box 397, Adelaide SA 5001, Australia b The University of Adelaide, PMB1, Glen Osmond, SA 5064, Australia c CCS Haryana Agricultural University, Hisar 125004, India d Formerly South Australian Research and Development Institute Staff, Australia e Department of Environmental Sciences, University of California, Riverside, CA 92521, United States f The University of New South Wales, Sydney, NSW 2052, Australia

ARTICLE INFO ABSTRACT

This manuscript was handled by Corrado Altering the soil surface features can potentially regulate water and solute movement processes in the soils, and Corradini, Editor-in-Chief, with the assistance reduce the accumulation of salts in the plant root zone. In this study, HYDRUS-2D was used consecutively for of Weiping Chen, Associate Editor three years (2011–2014) to evaluate the potential impact of different rainfall redirection and water harvesting Keywords: techniques such as no mounding or control (A), mid-row mounding (B), mid-row mounding covered with plastic Rainfall redirection (C), and plastic buried in the soil in mid-row (D) on water balance, root zone salinity dynamics, and salt balance Mulching in the soil in three grapevine producing regions (Loxton, McLaren Vale, and Padthaway) in South Australia. Leaching Simulations covered varied soil, climate, irrigation quality (0.3, 1.2, and 2.2 dS/m), and vine management Salinity conditions typical of each region. Seasonal transpiration (Tp), evaporation (Es), and drainage (Dr) accounted for Grapevine 39–49, 26–44, and 17–25%, respectively, of the total drip irrigation applied to the vineyards across the three HYDRUS-2D locations. Relative to the control, mounding mid-row soils (B) did not significantly alter the water balance atany

site; adding an impermeable plastic layer to that mound (C) reduced Es (48–54%), and increased Dr (by two

thirds) and Tp (1–16%). A tremendous ameliorative impact of mid-row mounding with plastic (C) was observed

in the spatiotemporal salinity dynamics in the soils at all three locations (salt removal efficiency LEs > 1). An increased leaching fraction under the mid-row mounding with plastic (C) led to considerable (up to 14.6 t/ha) salts leaching from the crop root zone, which was double the other treatments. Buried plastic (D) showed slightly better outcomes than the control or mid-row mounding, particularly for seasonal salt leaching in heavy textured

soils. Salt removal efficiency (LEs) > 1 in light- and medium-textured soils as compared to heavy-textured soils indicates better salt removal in the former soils. The results demonstrated that the mid-row mounding with plastic is an effective technique to reduce root zone salinity in the drip-irrigated horticultural crops.

1. Introduction that > 19 percent of the total irrigated land (about 62 m ha) in the world suffers from salinization. Every day for > 20 years, an averageof Irrigation induced soil salinization is a worldwide phenomenon, 2,000 ha of irrigated land in arid and semi-arid areas across 75 coun- especially in arid and semiarid regions where poor quality water is used tries have been degraded by salt-related degradation (Qadir et al., for crop production. It is not only impacting soil quality and reducing 2014). It is anticipated that increased temperature and reduced rainfall crop productivity but also curtailing the cultivated area and rendering projections by IPCC (2014) may intensify the risks of soil salinization the soil unfit for sustainable crop production (e.g., Cao et al., 2018). and endanger sustainable crop production, including high-value horti- Poor quality water includes untreated or recycled municipal sewage cultural crops such as grapevine. water, partially treated or untreated effluents/spent wash from in- Irrigated grapevines are grown across every continent and are dustries, groundwater and drainage water, which are often used for highly managed growing systems where water availability and its irrigating the crops as a supplement to the rainfall or a good quality quality are often the most limiting factors (Flexas, 2016; da Silva et al., source. Food and Agriculture Organization (FAO, 2016) reported 2018). In Australia, > 90% of grapevines are grown under highly

⁎ Corresponding author at: South Australian Research and Development Institute, GPO Box 397, Adelaide SA 5001, Australia. E-mail address: [email protected] (V. Phogat). https://doi.org/10.1016/j.jhydrol.2020.125000 Received 2 February 2020; Received in revised form 20 April 2020; Accepted 21 April 2020 Available online 23 April 2020 0022-1694/ Crown Copyright © 2020 Published by Elsevier B.V. All rights reserved. V. Phogat, et al. Journal of Hydrology 587 (2020) 125000 efficient micro-irrigation systems (ABS, 2015). South Australia is a (e.g., Liu et al., 2013; Wang et al., 2014; Li et al., 2015; Zhao et al., major grape and wine-producing region in Australia, contributing 48% 2018). These studies were only confined to evaluating the dynamics of of the total wine grape crush, and 93% of vineyards use supplementary water distribution in soils, and the reduction of evaporation due to drip irrigation (ABS, 2015). But, the quality of irrigation water used for mulched conditions on the flat surface by considering a no-flow grapevine cultivation is highly variable across different vine-growing boundary over the mulched surface. Simulations of water and salinity regions. For example, in McLaren Vale, recycled water (ECw = 1.0–1.7 dynamics under different mid-row rainwater harvesting techniques dS/m) is commonly used to supplement rainfall. In contrast, in the (e.g., mounding with and without plastic, and sub-surface plastic) South East region, groundwater (ECw = 1.9–2.4 dS/m) is the dominant coupled with drip irrigation of grapevines could potentially serve as a source for irrigating grapevines (Stevens et al., 2012; Pitt et al., 2015). new concept for implementing the surface boundary in modelling stu- Increased use of these waters (recycled and groundwater) for irrigation, dies. Furthermore, to our knowledge, no modelling study has been fo- frequent droughts, and climate change projections of diminishing pre- cussed on evaluating the ameliorative effect of rainfall directed from cipitation are putting enormous pressure on growers to maintain and the mid-row of horticultural crops. sustain viticulture production around the world (e.g., DeGaris et al., Therefore, in this investigation, HYDRUS-2D was used to evaluate 2015; van Leeuwen and Darriet, 2016). Besides, increased use of saline the impact of different rainfall redirection and water harvesting tech- water for irrigating grapevines is posing a potential danger of soil sal- niques (A - no mounding or control, B - mid-row mounding, C - mid-row inization, inflicting a serious impact on the long-term sustainability of mounding covered with plastic, and D - plastic buried in the soil in mid- these vineyards (Stevens et al., 2012). row) on water balance, root zone salinity dynamics, and salt balance in Irrigated agriculture has managed the use of saline waters by different grapevine growing regions (Loxton, McLaren Vale, and leaching of salts from the rootzone depending on crop salt tolerance Padthaway) over multiple seasons (2011–2014) involving varied soil, (Ayers and Westcot, 1985). The traditional irrigation management climate, irrigation quantity and quality, and crop conditions. The out- strategy under conditions where the supply of water is not limited is to come of this investigation would be helpful in evaluating these in- provide extra water such that the ratio of the actual depth of drainage novative practices for controlling soil salinization and salt leaching for to the depth of irrigation, i.e., the leaching fraction (LF), satisfies the sustainable grape production. leaching requirement (LR) (FAO, 1994). Due to the complexities of interactions between water, soil, and plant uptake, matching of LF with 2. Materials and methods LR is not straightforward (Dudley et al., 2008). In addition, the un- availability of good quality water and the complexity of implementa- 2.1. Description of the study sites tion of LR for salinity control under drip irrigation systems may lead to enormous amounts of salt deposition laterally in the soils (e.g., Hanson The present investigation was carried out in three grapevine et al., 2008). Complex soil heterogeneity and widespread stratification growing regions in South Australia (SA), i.e., McLaren Vale (McLaren in Australian grapevine growing areas) and poor drainage conditions Vale region), Padthaway (South East), and Loxton (Riverland) (Fig. 1). may also confound efforts to tackle soil salinization (Singh, 2019; These sites have different climate, soil, irrigation water quality, and Wichelns and Qadir, 2015). growing practices, representing diverse and varied conditions for Under such situations, on-farm water harvesting techniques such as grapevine cultivation. The water and salt balance modeling study was mulching and rainfall redirection play a key role in regulating water conducted for three consecutive seasons (2011–12 to 2013–14) at the and solute movement processes in the soil (Wang et al., 2009). For three locations. All experimental details including soil characterization, example, drip irrigation coupled with plastic mulch has been in- irrigation scheduling, water quality, crop performance and yield for troduced over a large area in China to reduce evaporation and to op- these trials can be found in Pitt et al. (2015), Stevens et al. (2012, timize the efficiency of available water resources (Liu et al., 2013). 2013), and Phogat et al. (2017, 2018b) for the McLaren Vale (McL), Several studies have reported benefits of mulching in term of water use Padthaway (Pad), and Loxton (Lox) sites, respectively. However, brief efficiency (Adhikari et al., 2016; Zhao et al., 2014; Yu et al., 2018), seed site-specific information about data acquisition relevant to the current emergence (e.g., Dong et al., 2010), crop yield (Qin et al., 2014; Yu study is given below. et al., 2018), salinity control (Bezborodov et al., 2010; Abd El-Mageed et al., 2016), soil temperature management (e.g., Li et al., 2017; Zhang 2.1.1. McLaren Vale et al., 2018), and weed control (e.g., Ramakrishna et al., 2006). Field McLaren Vale (McL) is a premium wine grape region in South experiments on rainfall redirection techniques (removing under vine Adelaide, which is characterized by a Mediterranean climate. The cli- mound and mounding in mid-row with rainfall harvesting) have es- mate data for the study site were obtained by running a data drill for tablished a reduction of the soil salinity and Na+ and Cl- content in the study site (Jeffrey et al., 2001). The average maximum temperature, wine grapes (Stevens et al., 2013; Pitt et al., 2015). However, it is not potential evapotranspiration, and rainfall for the last 100 years at the clear whether these techniques apply beyond the particular combina- experimental site were 20.9 °C, 1568.5 mm, and 555 mm, respectively. tions of water quality, climate, and soils at the experimental sites. In- Other data for simulations were obtained from an experimental study deed, it is worth examining the impact of these techniques on water conducted at a commercial Cabernet Sauvignon vineyard (35°14́S and balance, root zone salinity dynamics, and salt leaching in soils. 138°31́E) located within the region (Pitt et al., 2015). The vineyard was Conducting long-term experiments involving numerous variables planted in 1998 with a vine spacing of 2.75 m by 1.8 m. The vines were under varied soil, water, and climate conditions are costly, and labor- irrigated with a drip system with 1.2 L/h drippers spaced at 0.6 m and and time-intensive. On the other hand, numerical models are excellent aligned along the length of vine rows. Treated wastewater for irrigation cost-effective tools to study the impact of climate, soil, water, andcrop was obtained from the Christies Beach Wastewater Treatment Plant and variables on water and solute transport in the soil (Phogat et al., distributed through a pipeline scheme managed by Willunga Basin 2018a). Among the available agro-hydrological models, HYDRUS (2D/ Water (WBW). Water samples were collected from irrigation emitters 3D) (Šimůnek et al., 2016) has been widely used to simulate water throughout the growing season and assessed for salinity (Pitt et al., movement and salinity dynamics under drip irrigation (Ramos et al., 2015). The average irrigation water salinity (ECw) during the study 2012; Selim et al., 2013; Chen et al., 2015; Phogat et al., 2014, 2019) period (2011–14) was 1.2 dS/m (range 1.0–1.3 dS/m). The amounts of because of its flexibility to accommodate different types of complex seasonal supplementary irrigation applied to grapevine at McLaren Vale boundary conditions, to consider root water and nutrients uptake, and ranged from 86.6 to 135.3 mm during 2011–14 (Table 1). its ease of use due to a user-friendly graphical interface. However, few A comprehensive soil testing program was carried out at the modelling studies have included a mulched boundary for drip irrigation McLaren Vale study site due to the existence of a complex

2 V. Phogat, et al. Journal of Hydrology 587 (2020) 125000

Simpson Desert N Great Victoria Desert Lake Eyre

Lake Maurice

Lake Lake Frome Lake Torrens Gairdner Port Augusta

Adelaide Loxton Murray McLaren Vale River

Padthaway

Mt Gambier

Fig. 1. Locations of study sites (Loxton, McLaren Vale and Padthaway) in different grapevine regions in South Australia.

Table 1 are quite variable and are influenced by rainfall patterns and the types Annual rainfall (P) and seasonal irrigation (I) in vineyards during 2011–2014 at of irrigation practices. Surface drippers spaced at 60 cm were used to different locations in South Australia. apply supplementary groundwater irrigation to vines with a salinity Location 2011–12 2012–13 2013–14 Average varied from 1.9 to 2.3 dS/m. The average salinity of irrigation water PIPIPIPI (ECiw) over the three seasons (2011–14) was 2.2 dS/m. The pressure (mm) compensated drippers had a discharge of 2.1 L/h. In the absence of McLaren Vale 614.3 86.6 437.1 129.5 524.0 135.3 525.1 117.1 good quality surface water resources at Padthaway, irrigated vines were Padthaway 466.4 242.5 446.4 242.5 576.9 242.5 496.6 242.5 solely dependent on groundwater for supplemental irrigation. Seasonal Loxton 298.2 322.5 198.7 435.0 348.3 377.0 281.7 378.2 water applications during the simulation period amount to 242.5 mm (Table 1). The soils at the Padthaway study site varied from sandy clay loam at heterogeneous soil formation in this region (Pitt et al., 2015). Un- the surface to clay at lower depth (Stevens et al. (2012). The soil hy- disturbed core samples were collected from 0 to 15, 30–60, and draulic parameters (Table 2) used in the modelling study were esti- 60–100 cm soil depths under vine (UV), under the track (UT), and in mated from measured values of the water content-pressure head re- mid-row (MR) regions to examine the variability in soil characteristics lationship in undisturbed cores taken from 0 to 30 and 30–100 cm. (Pitt et al., 2015). The soil hydraulic parameters (Table 2) were esti- Other soil parameters (the saturated hydraulic conductivity and bulk mated following van Genuchten-Mualem constitutive relationship (van density) for both layers were also measured on undisturbed soil cores. Genuchten, 1980), utilizing measured soil water retention data on un- Similar hydraulic parameters were assumed at different locations across disturbed soil core samples collected from UV, UT, and MR regions. The the vine lines (UV, UT, and MR). While measured initial soil salinity saturated hydraulic conductivity (Ks) and the bulk density (Db) were (ECe) at UV down to the 80 cm depth varied from 4 to 4.5 dS/m at the also measured on undisturbed soil cores following standard procedures start of the simulation, the ECe at MR and UT (1.2 to 2.0 dS/m, re- (Klute, 1986). Hence, input parameters represent a complex variability spectively) were less than half of that (Stevens et al., 2012). existing at different soil depths and laterally across the vine rows.

2.1.3. Loxton 2.1.2. Padthaway Loxton (34°27́S and 140°34́E) is a part of the Riverland vineyard Padthaway (36°37́S and 140°32́E), in the South East of South region, which is located along the River Murray corridor and utilizes Australia, is characterized by the presence of medium-textured soils the good quality river water for grapevine irrigation. It has a warmer underlain with limestone rocks, popularly known as Terra rossa. Mediterranean climate and receives less rainfall compared to the other Padthaway (Pad) has a warm Mediterranean climate with good rainfall. two regions. The average maximum temperature, potential evapo- The average rainfall and potential evapotranspiration over the last transpiration, and rainfall for the last 100 years were 23.8 °C, 1823 mm, 100 years amount to 523.3 and 1553.4 mm, respectively. Climate and 265 mm, respectively (Phogat et al., 2017). parameters for modelling were taken from a nearby Bureau of The Chardonnay vineyard at Loxton (Lox) was irrigated with a Meteorology observatory. surface drip system. The water was applied through pressure compen- Relevant crop data for modelling was obtained from a field ex- sated drippers with a discharge of 1.6L/h. The plant spacing is similar periment in a commercial own-rooted Chardonnay vineyard, located to the vineyards at the other two locations. The vineyard was irrigated approximately 12 km south of Padthaway (Stevens et al., 2012, 2013). with water from the River Murray, which has salinity varying from 0.2 The vineyard was planted in 1996 with Chardonnay clone I10V1. to 0.4 dS/m, with a mean value of 0.3 dS/m during the study period Salinity trends in the shallow unconfined aquifer on the Padthaway Flat (Phogat et al., 2018b). Seasonal irrigation applications were much

3 V. Phogat, et al. Journal of Hydrology 587 (2020) 125000

Table 2 Estimated soil hydraulic parameters at different sites (McLaren Vale, Padthaway, and Loxton) used in simulations.

3 −3 3 −3 −1 -1 −3 Soil texture Soil depth (cm) θr(cm cm ) θs(cm cm ) α(cm ) n Ks(cm d ) l Db(g cm )

McLaren Vale Sandy clay loam (UV) 0–30 0.23 0.50 0.014 1.5 137.8 0.5 1.50 Sandy clay loam (UT) 0–30 0.23 0.46 0.013 1.5 57.4 0.5 1.55 Sandy clay loam (MR) 0–30 0.19 0.47 0.01 1.4 153.9 0.5 1.44 Clay (UV) 30–65 0.26 0.49 0.014 1.4 20.7 0.5 1.55 Clay (UT) 30–65 0.31 0.49 0.03 1.4 33.2 0.5 1.55 Clay (MR) 30–65 0.27 0.50 0.01 1.4 87.7 0.5 1.40 Clay (UV) 65–100 0.26 0.49 0.114 1.2 58.7 0.5 1.49 Clay (UT) 65–100 0.27 0.46 0.1 1.2 13.8 0.5 1.49 Clay (MR) 65–100 0.25 0.43 0.014 1.4 35.1 0.5 1.32 Padthaway Sandy clay loam 0–30 0.07 0.48 0.020 1.25 26.0 0.5 1.3 Clay loam 30–100 0.17 0.44 0.058 1.3 11.0 0.5 1.44 Loxton Loamy sand 0–30 0.04 0.4 0.027 2.2 388.8 0.5 1.53 Loamy sand 30–60 0.05 0.38 0.04 1.7 259.2 0.5 1.41 Loam 60–100 0.05 0.37 0.04 1.62 172.8 0.5 1.44

θr is the residual water content, θs is the saturated water content, Ks is the saturated hydraulic conductivity, Db is the bulk density, and α, n, and l are the van Genuchten shape parameters. higher at Loxton as compared to the other two locations due to less heads at the flux boundary could make the numerical code unstable rainfall and low water retention capacity of light-textured soils. The (Šimůnek et al., 2016). During irrigation, the drip line boundary was average seasonal irrigation over the three seasons at Loxton was held at a constant water flux, q, equal to the dripper discharge rates at 378 mm (Table 1). different locations. The atmospheric boundary condition was assumed The soils at the Loxton site are predominately sandy in texture. The for the remainder of the soil surface. A no-flow boundary condition was soil hydraulic parameters for this modelling study were taken from specified on the left and right edges of the soil profile to account for Phogat et al. (2017) and are shown in Table 2. The saturated hydraulic flow and transport symmetry. A free drainage boundary condition was conductivity (Ks) and the bulk density (Db) were measured on un- assumed at the bottom of the soil profile. disturbed core samples using standard procedures. Similar soil hy- To implement a plastic cover over the mound (treatment C), arcs draulic parameters were assumed at different locations across the vine were constructed 3–5 cm below the surface boundary, and a no-flow lines [under vine (UV), under the track (UT) and in the mid-row (MR) boundary condition was imposed in the domain over the plastic sheet regions]. Furthermore, HYDRUS-2D was calibrated and validated for on both sides at the desired locations (see Fig. 3). The extent of eva- spatial and temporal distributions of the soil water content over mul- poration from the soil over the no flow mulched surface was assumed tiple seasons (Phogat et al., 2017). equal to the losses due to wear and tear of the mulch. Liu et al. (2013) reported that plastic mulch reduced the soil evaporation by 90% under optimum conditions. However, year-old mulch can deteriorate and be 2.2. Modelling domain and boundary conditions for different treatments torn apart at some places, leading to significant evaporation losses. Essentially, placing the plastic mulch below a thin soil layer helped to Field experiments at Padthaway and McLaren Vale consisted of four protect it from the hot weather and other mid-row operations (Pitt rainfall harvesting and redirection treatments, i.e., control (A), a mid- et al., 2015). Another advantage of adopting this approach was to fa- row mound (B), a mid-row mound covered with plastic (C), and buried cilitate rainwater movement along the ridge/mound to the lower plastic in the mid-row region (D). The establishment of these treatments reaches of the furrow/soil. This approach, further, allowed simulating at field sites is discussed in Stevens et al. (2012) and Pitt et al. (2015), the variable impact of dynamic pressure heads due to rainfall on flow respectively, and is shown in Fig. 2 for the McLaren Vale site. Similar conditions in the soils. Similarly, for treatment D, a no-flow boundary modelling scenarios were conducted at Loxton to match the treatments was embedded into the domain at the location (10 cm below the soil at the other two sites. In the control treatment, actual mid-row condi- surface) where the plastic sheet was placed in the field studies. Fine tions show depressions in the wheel ruts due to heavy traffic associated mesh size was adapted below the time-variable flux boundary (below with mechanical pruning (or pre-pruning), fungicide/herbicide spray the dripper) in all treatments to facilitate dynamic flow conditions in program, and mechanical harvest operations. In the other treatments response to frequent irrigations. All specified boundary conditions are (B, C, and D), a shallow ripping operation and/or removing the under illustrated in Fig. 3. Simulations were conducted separately for each vine mounding soil facilitated the mid-row mounding operations. These treatment for three years (July 1, 2011- June 30, 2014) using daily time modifications on the soil surface are also illustrated in the modelling steps for boundary conditions. domains (Fig. 3). The modelling domains for different treatments were constructed in

HYDRUS-2D (Šimůnek et al., 2016) based on the vine spacing, drip 2.3. Estimation of potential transpiration (Tp) and potential evaporation design parameters, and mid-row treatments (Fig. 3). The vine row is (Es) present in the centre of the domain, which was extended equally on both sides across the vine row. The vertical depth was equal to 100 cm, HYDRUS-2D requires daily values of potential Tp and Es under the which relates to the rooting depth of grapevines in the field experiments field conditions as inputs and then calculates modelled values de- (Stevens et al., 2012; Pitt et al., 2015; Phogat et al., 2017). A time- pending on soil moisture dynamics. In this study, the FAO-56 dual-crop variable flux boundary condition was applied to a 20 cm long boundary coefficient approach was employed to estimate the daily values ofpo- directly below the dripper, centred on 137.5 cm from the top left corner tential Tp and Es for grapevine at three locations (Allen et al., 1998). of the soil domain (Fig. 3). The length of this boundary was selected to This method has been used previously in many studies for input data ensure that all irrigation water could infiltrate into the soil without generation (e.g., Phogat et al., 2014, 2017). It combines two coeffi- producing positive surface pressure heads, because positive pressure cients, i.e., a basal crop coefficient (Kcb) responsible for transpiration

4 V. Phogat, et al. Journal of Hydrology 587 (2020) 125000

Fig. 2. Field view of the rainfall redirection treatments (A-control, B-Mid-row mound, C-Mid-row mound with plastic, and D-Buried plastic) imposed at the McLaren Vale experimental site.

Fig. 3. Schematic view of the HYDRUS-2D spatial modelling domain for A) control, B) mid-row mound, C) mid-row mound with plastic mulch, and D) buried plastic treatments based on vine and drip spacing and mid-row dynamics. Applied boundary conditions are also shown.

and a soil evaporation coefficient (Ke) (i.e., Kc = Kcb + Ke), with daily fertigation management. The longitudinal dispersivity was assumed as reference crop evapotranspiration (ET0) to estimate daily crop evapo- one/tenth of the modelling domain (with the transverse dispersivity transpiration (ETC) as follows: being one-tenth of the longitudinal dispersivity) (Beven et al., 1993; Cote et al., 2003) and the molecular diffusion coefficient of salts in ETC=() K cb + K e ET0 (1) water was considered as 1.656 cm2/day (Phogat et al., 2017). The data Standard Kcb values for grapevine were adjusted for the local cli- for EC of irrigation water (ECiw) was based on the water quality ana- mate, taking into consideration crop height, wind speed, and minimum lysis. However, average values of ECiw were considered for modelling relative humidity averages for the period under consideration for all locations (McLaren Vale = 1.2, Padthaway = 2.2, and (2011–2014) at all locations (Allen et al., 1998). Location-specific Loxton = 0.3 dS/m) as there was only a small variation in the irrigation parameters used to estimate daily Es, and Tp for grapevine following the water quality over the study period. The rainfall chemistry analyzed by FAO-56 dual-crop coefficient approach for three locations are given in Cresswell et al. (2010) in the Adelaide region estimated the salinity

Table 3. Values of daily potential transpiration (Tp) and soil evaporation (ECrw) of 0.12 dS/m, and this value was used in simulations as a (Es) obtained in this way were used as time-variable boundary condi- common value for all locations. To facilitate the prediction of the tions (see Fig. 3) in the HYDRUS-2D model, along with precipitation leaching of salts from the soil profile, all input ECiw and ECrw values received at the site during the study period. More details about the were converted into mass units using a common approximation (1 EC FAO-56 dual-crop coefficient can be found in Allen et al. (1998) and (dS/m) = 640 mg salts/L). The salt leaching efficiency was estimated to Allen and Pereira (2009). compare the effectiveness of various rainfall redirection techniques. It was estimated for all locations annually as the amount of salts leached 2.4. Solute parameters for salinity dynamics in the soil below a 1 m depth per volume of water drained. The initial water content distribution was set to either measured

Soil solution salinity (ECsw) distribution was modelled as a non-re- values (McLaren Vale and Padthaway) or to field capacity (Loxton). active solute (e.g., Ramos et al., 2011; Wang et al., 2014; Phogat et al., Measured values of ECe in the soil were converted to salinity at actual soil 2014, 2019). These studies demonstrated that this approach could be water content following a linear relationship by Pitt et al. (2015) and to successfully used in environments under intensive irrigation and salt mass as per the approximation 1 EC (dS/m) = 640 mg salts/L.

5 V. Phogat, et al. Journal of Hydrology 587 (2020) 125000

Table 3 for buried plastic (D) relative to the control. The mid-row Values of important parameters used to estimate daily transpiration (Tp) and mound + plastic (C) maximised vine water uptake followed by the daily evaporation (Es) for vineyards at different locations. buried plastic treatment. The average increase in grapevine Tp in the Parameters McLaren Vale Padthaway Loxton mid-row mound + plastic treatment was 16% higher than the control. Normally, transpiration/root water uptake are expected to increase for Kcb ini 0.20 0.20 0.20 sites with plastic mulch as compared to no mulch sites (Allen et al., K 0.65 0.65 0.65 cb mid 1998), because temperatures are higher under mulch, and root growth Kcb end 0.5 0.5 0.5

Kcb mid adj (2011–12) 0.66 0.72 0.71 is promoted (Gao et al., 2014; Saglam et al., 2017). At the same time, Kcb mid adj (2012–13) 0.67 0.71 0.73 there was a drastic reduction (54%) in Es in the mid-row Kcb mid adj (2013–14) 0.69 0.72 0.72 mound + plastic treatment, which may have produced higher Dr, Kcb end adj (2011–12) 0.51 0.57 0.56 especially during the 2012–13 and 2013–14 seasons. Therefore, the Kcb end adj (2012–13) 0.52 0.56 0.58 average seasonal Dr in mid-row mound + plastic treatment increased Kcb end adj (2013–14) 0.54 0.57 0.57

fw, surface drip 0.29 0.39 0.22 three times compared to the other treatments. Moreover, the average fw, rain 1 1 1 soil water storage (ΔS) also increased about four times under mid-row RAW (mm) 74 46 34 mound + plastic as compared to the control. The water conservation TAW (mm) 150 85 60 effect of plastic mulch has been documented in many previous studies TEW (mm) 20 20 12 REW (mm) 10 10 8 for drip irrigation (e.g., Saglam et al., 2017). Plant height h (m) 1.5 1.5 1.5 The average water balance components at Padthaway were com- Rooting depth Zr (m) 1.0 1.0 1.0 parable to the McLaren Vale site. Nonetheless, the average soil water Vineyard density (Fc eff) 0.5 0.5 0.5 storage (ΔS) at the end of the grapevine season was 5% higher, and the Evaporable depth Ze (m) 0.15 0.15 0.15 Mid-season Min RH (%) (2011–12) 40 36 30 average Es component was 7% lower than at the McLaren Vale site Mid-season Min RH (%) (2012–13) 36 33 26 (Table S-1, supplementary material). Similarly, Tp and Dr also increased Mid-season Min RH (%) (2013–14) 37 34 27 by 7.5% and 19% (almost three times), respectively, when the control Av. Wind speed (m/s) (2011–12) 1.77 3.06 2.16 treatments were compared. On the other hand, at Loxton, rainfall re- Av. Wind speed (m/s) (2012–13) 1.41 2.37 2.24 direction techniques had a little impact on vine water uptake as T Av. Wind speed (m/s) (2013–14) 2.43 2.83 2.17 p remained similar in all treatments (Table S-2, supplementary material).

Kcb are the basal crop coefficients for initial (ini), mid (mid), and late (end) seasons However, average Tp accounted for 49% of the applied water, while adjusted (adj) to local climate, fw is the wetted soil surface fraction, RAW is 26% of irrigation and rainfall contributed towards Es, with the re- readily available water, TAW is total available water, TEW is total evaporable maining 25% of water draining out of the soil profile (1 m). Drainage water, and REW is readily evaporable water. (Dr) almost doubled in the mid-row mound + plastic treatment (C) compared to the control (A), showing the impact of surface mulching on 3. Results and discussion direct water loss from the soil surface. Leaching fractions (LF) estimated for the range of treatments at the 3.1. Seasonal water balance three locations are shown in Fig. 4. The average LF for the mounding only (B) and buried plastic (D) treatments were similar to the control Seasonal water balance components for grapevine under different (A). However, annual LFs were influenced by the site-specific soil tex- rainfall redirection treatments at the McLaren Vale site are shown in ture and climate conditions (rainfall), being higher at Loxton Table 4. The average water balance, irrespective of seasons and treat- (0.15–0.25). Notably, LFs were considerably lower at Padthaway ments, divided the transpiration (Tp), evaporation (Es), and drainage during 2011–12 due to a relatively dry soil profile at the start of the (Dr) as 39, 44, and 17% of the total water application, respectively. simulation compared to the other sites. Later on, the LF at Padthaway Among the treatments, the results were similar for the mid-row matched corresponding values at McLaren Vale. On the other hand, the mounding treatment (B) and the control (A), while Tp increased slightly LF in the mid-row mound + plastic (C) treatment increased on average three times at McLaren Vale (0.26–0.39) and Padthaway (0.16–0.4) and two times at Loxton (0.31–0.4) compared to the corresponding average Table 4 values for other treatments. Hanson et al. (2008) reported similar Simulated components of the annual water balance for vineyards under dif- leaching fractions (0.07–0.31) under drip irrigation, which encourages ferent rainfall redirection treatments for three consecutive seasons (2011–2014) at McLaren Vale. rapid localized leaching, especially in light-textured soils. Indeed, treatment mid-row mound + plastic has emerged as the Year/ Treatment IPTp Es Dr ΔS mm 0.4 A) Control 2011–12 82.16 614.3 245.14 340.90 65.91 51.21 McL Pad Lox 0.3 2012–13 129.44 437.1 196.37 317.20 85.53 −25.00 2013–14 135.20 524.0 261.66 309.28 93.96 −14.40 B) Mid-row mound 0.2 2011–12 82.16 614.3 245.94 341.79 63.96 52.21 2012–13 129.44 437.1 195.04 318.78 86.31 −26.47 0.1 2013–14 135.20 524.0 259.14 310.05 94.73 −13.43 Leaching fraction C) Mid-row mound + plastic 2011–12 82.16 614.3 289.89 155.31 183.98 79.97 0.0 2012–13 129.44 437.1 231.11 147.53 219.92 −25.89 2013–14 135.20 524.0 295.07 142.07 219.55 −6.98

D) Buried plastic 2011-12 2012-13 2013-14 2011-12 2012-13 2013-14 2011-12 2012-13 2013-14 2011-12 2012-13 2013-14 2011–12 82.16 614.3 264.68 343.43 28.57 58.14 2012–13 129.44 437.1 207.57 319.17 65.69 −14.41 A B C D 2013–14 135.20 524.0 277.11 307.34 90.17 −18.47 Fig. 4. Annual leaching fractions under grapevine for different treatments (A- control, B- Mid-row mound, C- Mid-row mound with a plastic cover, D- Buried Tp is seasonal transpiration; Es is seasonal evaporation; Dr is seasonal drainage; ΔS is soil storage/depletion plastic) at McLaren Vale (McL), Padthaway (Pad), and Loxton (Lox).

6 V. Phogat, et al. Journal of Hydrology 587 (2020) 125000

i) McLaren Vale 2011-12 2012-13 2013-14

A A A

B B B Soil depth (cm) C C C

D D D

ii) Padthaway

A A A

B B B Soil depth (cm) C C C

D D D

iii) Loxton

A A A

B B B

Soil depth (cm) C C C

D D D

Lateral distance from the vine (cm)

Fig. 5. Electrical conductivity (ECe) distributions in the soil at i) McLaren Vale, ii) Padthaway, and iii) Loxton in different treatments (A - control, B - Mid-row mound, C - Mid-row mound + plastic, D - Buried plastic) at the end of 2011–12, 2012–13, and 2013–14 grapevine seasons. most favourable treatment with increased plant water uptake, a higher 3.2. Soil salinity dynamics fraction of the evaporative flux diverted towards leaching, a substantial increase in the LF, which altogether can help in maintaining a favour- 3.2.1. Spatiotemporal salinity distribution in the soil able environment by transporting salts out of the root zone. These re- The impact of the rainfall redirection techniques on spatial salinity sults further confirm that plastic mulching, compared to the mid-row (ECe) dynamics in the soil at the end of 2011–12, 2012–13, and 2013–14 mounding, was an effective on-farm water management strategy that growing seasons at the three locations is shown in Fig. 5. Generally, the can improve the rhizosphere environment for a long-term sustainable ECe distribution at the McLaren Vale and Padthaway sites in the under- grapevine production. vine (UV) was higher than in the mid-row irrespective of the treatment because drip irrigation applies water close to the vines. Rapid water

7 V. Phogat, et al. Journal of Hydrology 587 (2020) 125000

8 8 Fig. 6. Temporal changes in the average A) Control UV_M MR_M B) MR mound profile ECe (dS/m) in the soil as influenced 6 6 by different treatments (A - control, B -Mid- )m/Sd( UV_P MR_P row (MR) mound, C - MR mound + plastic, D - Buried plastic) under the vine (UV) and 4 4

e in the mid-row (MR) region at the McLaren Vale (UV_M; MR_M) and Padthaway (UV_P EC 2 2 and MR_P) sites during July 2011- June 2014. 0 0 8 8 C) MR mound+plastic D) Buried plastic )m/Sd( 6 6

e 4 4 EC 2 2

0 0

extraction due to the high root density in this region results in the de- treatment mid-row mound + plastic (C) continued to be by far the most position of salts. On the other hand, in the mid-row, reduced water up- effective at salt removal compared to the other treatments. take and little salt addition via irrigation coupled with adequate rainfall- Interestingly, at Loxton, ECe remained lower in the UV region as induced salt flushing produced lower ECe. However, at McLaren Vale, a compared to MR, irrespective of the rainfall redirection treatments trend towards increasing ECe occurred, especially in the under-vine at the (Fig. 5iii). At the end of the 3rd season (2013–14), a low ECe zone end of successive seasons, which is correlated with proportionally higher (< 0.5 dS/m) increased laterally in all treatments. The presence of additions of salts through irrigation coupled with less localized leaching. sandy soils and a three times higher amount of irrigation water of ECiw Among the treatments, the bigger ECe “bulge” was present in the under- of 0.3 dS/m applied within a localized region facilitated rapid leaching vine in the control (A) and the mid-row mound (B) treatments as com- of the salts. In contrast, ECiw at McLaren Vale and Padthaway was 4 and pared to the mid-row mound + plastic(C). At the end of the 2013–14 7 times higher, respectively, than at Loxton, which potentially added an season, the maximum ECe in the under vine region was 4.2, 3.5, 3.0, and enormous quantity of salts in the soil in the under-vine area in these 3.5 dS/m, respectively, in the control (A), a mid-row mound (B), a mid- drip-irrigated vineyards. ECe in the soil profile always remained below row mound + plastic (C), and buried plastic (D) treatments at McLaren the grapevine threshold (2.2 dS/m) in all treatments. The buried plastic Vale. However, salinity remained below 0.5 dS/m in a mid-row treatment allowed a small amount of salt deposition just below the mound + plastic (C), whereas it was < 0.9 dS/m in the control (A) and plastic; this is due to the plastic membrane blocking vertical drainage of mid-row mound (B) treatments and much higher in the buried plastic (D) water and promoting lateral salt migration to the area underneath the (0.9–1.5 dS/m). The results showed that at McLaren Vale, the pre- plastic. In mid-row mound + plastic (C), the low salinity zone (< 0.5 dominance of heavy-textured heterogeneous soils and high irrigation dS/m) extended further under the vine compared to the other treat- water salinity encouraged salt deposition in the vicinity of the vine. This ments. It is also evident that most of the irrigation-induced salts were situation is similar to the poorly drained soils, which potentially en- leached out of the 1 m soil profile. Hence, the current irrigation amount courage salinization and waterlogging problems in irrigated agro-systems and quality of water did not pose any salinity threat in the root zone of (Wichelns and Qadir, 2015; Singh, 2019). the vine, irrespective of treatments. Christen et al. (2007) showed a

At Padthaway, the extent of ECe distribution in the soil was much similar impact on the salinity distribution when using good quality ir- greater than at McLaren Vale, irrespective of regions and treatments, rigation water in a loamy soil. because the initial salinity in the under-vine (4–4.5 dS/m) and mid-row Daily profile-averaged ECe in the under-vine and mid-row regions at (1.2–1.8 dS/m) regions was very high compared to the other two sites. the McLaren Vale and Padthaway sites are shown in Fig. 6. Data for the

Apparently, the high ECiw of the irrigation water (2.0–2.35 dS/m), cou- Loxton site is not shown as the salinity levels were much lower than the pled with medium to heavy-textured soils with low hydraulic con- grapevine threshold ECe of 2.2 dS/m (Zhang et al., 2002). A progressive ductivity (11–26 cm/day), is conducive to salt deposition within the soil increase in ECe was found at McLaren Vale in all treatments over the profile in the immediate vicinity of the vine (under-vine region), where three seasons except in mid-row mound + plastic (C) where ECe re- irrigation is applied through drippers. Therefore, pockets of high ECe mained almost constant (< 0.5 dS/m). ECe (0.9 dS/m) in buried plastic (5.5–7.0 dS/m) at Padthaway still existed under the vine below the 50 cm (D) was maximum in the mid-row region at the end of the simulation. depth in all treatments except mid-row mound + plastic at the end of the Despite increasing trends, ECe remained below the crop threshold in the 2011–12 vine growing season (Fig. 5ii). The maximum salinity zone had mid-row region in all treatments. On the other hand, it increased above an average ECe from 3.5 to 4.0 dS/m at the end of 2013–14. However, the threshold in the under-vine during the latter half of 2013–14, ir- this zone spreads over a large area in the control (A) and mod-row mound respective of treatments. It seems that heavy-textured soils with en- (B) treatments as compared to buried plastic (D), where it was restricted ormous profile heterogeneity and anisotropic flow conditions (Pitt within a small region, mostly in the deeper zone of the profile. In mid- et al., 2015) have played a crucial role in nullifying the treatment im- row mound + plastic (C), a significant volume of water passed through pact at McLaren Vale compared to the other Padthaway. the zone of high salinity, pushing the salts out of the profile. Conse- At the Padthaway site, the average ECe in the MR region remained quently, salinity was greatly reduced during the first season. During the below the threshold in all treatments except buried plastic (D). following seasons (2012–13; 2013–14), the soil profile salinity was re- Interestingly, in this treatment, ECe under the mid-row increased above duced below threshold (2.2 dS/m) in almost the entire mid-row, whereas the threshold during the 2013–14 growing season. Buried plastic pos- pockets of higher salinity existed in the under-vine below 30 cm. Hence, sibly harboured more salts directly below the plastic layer due to the

8 V. Phogat, et al. Journal of Hydrology 587 (2020) 125000 blockage of vertical drainage, as was also seen at Loxton also (Fig. 5). ha), and Loxton (136 to 247 kg/ha) were much lower as compared to However, the average ECe in the under-vine reduced drastically in all salts added through irrigation during the same period. treatments after an initial increase during the 2011–12 summers. Al- Although salt additions in the soil in different treatments were si- though it remained higher than the grapevine threshold (2.2 dS/m), the milar, salt storage (ΔS_S) and leaching (Dr_S) altered due to different impact of the rainfall redirection was very dramatic in the mid-row rainfall redirection techniques (Table 5). Interestingly, all treatments mound + plastic (C) treatment where ECe fell below the threshold showed an increased amount of salt remaining in the profile at McLaren during the post-winter season of 2013–14. The mean ECe in the buried Vale and Padthaway (except mid-row mound + plastic) during all plastic (D) treatment falls between values obtained in the control (A) seasons where recycled water and groundwater were used for irriga- and mid-row mound + plastic (C) treatments in the under-vine area. In tion, respectively. On the other hand, at Loxton, salt leaching was very other studies also mulching has been found to be a better field-man- rapid, and a high amount of leaching occurred compared to the amount agement option to reduce the upward movement of salts and eva- added through irrigation and rainfall (Table 5). Therefore, there was a porative water losses (Zhao et al., 2014; Chen et al., 2016) and that net depletion of salts from the soil at the end of the 2013–14 season in mulching with different materials is a promising technique for salinity all treatments. Similarly, the average salts storage at Padthaway in the control in agriculture (Bezborodov et al., 2010; Abd El-Mageed et al., mid-row mound + plastic (C) treatment was reduced by half as com- 2016). pared to the other treatments due to a rapid increase in the drainage The mid-row mound + plastic treatment (C) maintained relatively flux, which, however, was not sufficient to flush all salts out of thesoil lower salinity within the vine root zone as compared to the other profile. This suggests that soil texture and heterogeneity play anim- treatments, especially for recycled water and groundwater irrigated portant role in the transient salt storage in the soil. grapevines with heavier-textured soils, providing a better growing en- The average annual amount of salts leaching (Dr_S) in the control vironment for vine and a superior salt leaching approach as compared treatment (A) at each of the study sites remained almost similar to the to other techniques. The results of the modelling suggest that the mid-row mounding (B) over the three seasons (Table 5). Similarly, the treatments behaved differently at different locations, and the impact on dynamics of daily salts leaching in these treatments (A and B) almost the ECe distribution under grapevine due to rainfall redirection tech- overlapped over the entire simulation period (2011–2014) at all loca- niques depends on combined effects of the soil type, water quality, and tions except for slight deviations at Loxton (Fig. 6). This suggests that weather conditions. mid-row mounding (B) had a small impact on salts leaching from the root zone, irrespective of soil, water, and climate conditions. However, 3.3. Seasonal salt balance the mid-row mound + plastic (C) treatment resulted in more than twice the average salt mass removed (884 kg/ha) as compared to the control Salt balance dynamics in the soil under different rainfall redirection (406 kg/ha) at McLaren Vale. Similarly, the buried plastic (D) treat- techniques is shown in Table 5. There were different initial amounts of ment also showed higher average salts leaching (563 kg/ha) than the salts in the soils at different sites due to textural differences, andthe control (A) at Loxton. Although the average amount (4860 kg/ha) of quantity and quality of water used for grape production. At Padthaway, salts leached (Dr_S) at Padthaway in the mid-row mound + plastic initial salt contents (8298 kg/ha) in the soil profile were 3.5 and 5 times treatment (C) was more than one and half times of the control (A), it higher than at McLaren Vale and Loxton, respectively. The annual was about five times the corresponding amount at McLaren Vale. Itis amount of salts added through irrigation (I_S) during the simulation estimated that about 14.6 t/ha salts were leached during 2011–14 at period (2011 to 2014) varied from 631 to 1038 and from 900 to Padthaway in the mid-row mound + plastic treatment (C), which was 1200 kg/ha at the McLaren Vale and Loxton sites, respectively. How- 57% higher than for the control. Total salt leaching in other treatments ever, almost similar amounts of I_S (3352.8 kg/ha) were added annually ranged from 9.3 to 9.7 t/ha. The drastic increase in the salt leaching in at Padthaway during different years due to less variation in the seasonal the mid-row mound + plastic (C) treatment occurred due to the two- irrigation volumes. Additionally, the amount of salts added through the pronged strategy of the redirection of rainfall and the reduction of rain (P_S) at McLaren Vale (302–434 kg/ha), Padthaway (308–408 kg/ evaporation losses due to the surface cover. It can be seen that the

Table 5 Seasonal salt additions by irrigation (I_S), precipitation (P_S), salt leaching (Dr_S), and soil storage/ depletion (ΔS_S) in the soils (kg/ha) for grapevine at different locations under different treatments during 2011–2014.

Treatments Year McLaren Vale Padthaway Loxton I_S P_S Dr_S ΔS_S I_S P_S Dr_S ΔS_S I_S P_S Dr_S ΔS_S

A Initial salts 2394.4 8297.5 1618.5 2011–12 631.0 433.9 292.7 3166.6 3352.9 330.4 427.3 11553.4 901.5 209.6 1396.6 1333.1 2012–13 994.1 302.3 377.6 4085.3 3352.9 308.0 3158.0 12056.4 1201.6 132.7 1315.8 1351.5 2013–14 1038.4 373.7 548.0 4949.4 3352.7 408.3 5671.3 10146.0 1069.9 248.0 1857.4 812.0 Average 887.8 370.0 406.1 4067.1 3352.8 348.9 3085.6 11251.9 1057.6 196.8 1523.3 1165.5 B Initial salts 2454.1 8702.5 1635.0 2011–12 631.0 433.9 287.9 3231.1 3352.9 330.4 465.0 11920.9 901.5 208.4 1381.9 1362.9 2012–13 994.1 302.3 408.9 4118.5 3352.9 308.0 3012.8 12569.0 1201.6 132.7 1366.5 1330.7 2013–14 1038.4 373.7 591.1 4939.5 3353.1 408.3 5848.4 10482.0 1069.9 248.0 1763.2 885.4 Average 887.8 370.0 429.3 4096.4 3353.0 348.9 3108.7 11657.3 1057.6 196.4 1503.9 1193.0 C Initial salts 2440.9 8816.5 1618.0 2011–12 631.0 433.9 840.2 2665.6 3352.9 330.4 3828.0 8671.8 901.5 209.6 1957.3 771.8 2012–13 994.1 302.3 839.7 3122.3 3352.9 308.1 6648.0 5684.7 1201.6 132.7 1370.2 735.9 2013–14 1038.4 373.7 971.6 3562.8 3352.7 408.2 4104.7 5341.0 1069.9 248.0 1441.6 612.2 Average 887.8 370.0 883.8 3116.9 3352.8 348.9 4860.2 6565.8 1057.6 196.8 1589.7 706.6 D Initial salts 2375.1 8122.5 1605.2 2011–12 631.0 433.9 246.6 3193.4 3352.9 330.4 440.3 11365.5 901.5 209.6 1208.8 1507.6 2012–13 994.1 302.3 495.1 3994.7 3352.9 308.1 4999.7 10026.8 1201.5 132.7 1146.3 1695.5 2013–14 1038.4 373.7 948.1 4458.7 3353.1 408.2 4306.9 9481.2 1069.9 248.0 1558.0 1455.4 Average 887.8 370.0 563.2 3882.3 3353.0 348.9 3249.0 10291.1 1057.6 196.8 1304.4 1552.8

A is control, B is Mid-row mound, C is Mid-row mound + plastic, D is Buried plastic

9 V. Phogat, et al. Journal of Hydrology 587 (2020) 125000 drainage flux (6.7 ML/ha) in this treatment was more than doubleas 2.4 y (Lox) = -5.835 x2 + 3.18 x+ 0.79 compared to the other treatments (2.6–2.7 ML/ha) (Table 4). Differ- 2.0 ences in daily salt leaching in different treatments and at different lo- y (Pad) = 5.599 x0.988 cations are also distinctly visible during the entire simulation period 1.6 McL

(Fig. 6). High episodic salts leaching peaks occurred especially at the sEL Pad McLaren Vale and Padthaway sites, whereas leaching was a continuous 1.2 process in the sandy soil at Loxton. Typically, large peaks at all loca- 0.8 Lox tions were preceded by high rainfall events and were also influenced by the amount of salts present in the soil. Notably, at Padthaway, huge 0.4 y (Mcl) = 1.195 x0.56 amounts of salts (3.35 t/ha/yr) are added in the soils (see Table 5) 0.0 through high salinity (2.2–2.35 dS/m) groundwater irrigation, which 0.00 0.10 0.20 0.30 0.40 0.50 are deposited in the soil and ready to be transported following large LF rainfall events generating large leaching fractions. Indeed, the huge amount of salt leaching at Padthaway, irrespective Fig. 8. Relationship between the salt leaching efficiency (LEs) (kg salts leached/ of treatments, may raise salinity concerns for shallow groundwater kg salts added) and a leaching fraction (LF) in different soils at McLaren Vale (McL), Padthaway (Pad), and Loxton (Lox). aquifers. A groundwater monitoring study observed an increase in the salt contents by about 600 mg/L in shallow aquifers from around 800 mg/L to 1400 mg/L in parts of Padthaway during the last 25 years treatments and seasons, indicating a strong depletion pattern. However, (Cleugh, 2006). Apparently, excessive transport of irrigation-induced the mid-row mound + plastic (C) treatment had a maximum LEs value root zone salts under grapevine may pose a potential danger for the of 1.74 kg salts/kg salts applied during 2011–12. Hence, the occurrence groundwater quality. Hence, a trade-off exists between the extent of of LEs larger than one at Loxton is attributed to the prevalence of highly leaching and the imminent threat of groundwater degradation and, permeable sandy soils, low salinity of irrigation water (0.3 dS/m), and judicious decisions are required for sustainable salinity management almost double amount of irrigation application than at the other sites, and controlling the pollution of groundwater. which facilitated the rapid flushing of salts from the root zone. In drip-irrigated systems, it is hard to determine the leaching requirement (LR) due to spatially-variable soil wetting patterns that lead 3.4. Leaching efficiency and leaching requirement to localized leaching below the drip line (Hanson et al., 2008). His- torically, LR is the LF, which when passed through the root zone, re-

Salt removal (leaching) efficiency (LEs) is defined as the ratio of the duces the root zone salts below the crop threshold. Hence, in drip ir- amount of salts leached from the soil profile and the corresponding rigation systems, a LF that corresponds to LEs ≥ 1 can potentially leach amount added during a given time (e.g., an annual or crop cycle), annually added salts from the root zone. It can serve as a good estimate which offers a more quantitative measure of the salt balance to compare for leaching of irrigation induced salts from the root zone. Hence, a site- different treatments (Lu et al., 2019). At McLaren Vale, the LEs re- specific relationship was developed between a seasonal LF and corre- mained less than one irrespective of treatments and seasons, indicating sponding LEs at different locations, irrespective of treatments (Fig. 8). an overall tendency of salt deposition in the root zone (Fig. 7). How- At Lox (sandy soils), a very low LF (0.1) was found to rapidly attain ever, LEs increased almost two-fold in the mid-row mound + plastic (C) LEs > 1, while at Padthaway (medium-textured soils), the threshold LF treatment compared to the control and mid-row mounding, and varied was 0.18. The threshold LF for the McLaren Vale site (heavy textured, from 0.63 to 0.77 kg salts leached/kg added. Interestingly, at Padth- heterogeneous soils) was exceptionally high (0.74), which seems im- away, during 2011–12, LEs was very low (average 0.12–0.13 kg lea- practical. However, in the mid-row mound + plastic treatment (C), an ched/kg salts added) in all of the treatments besides mid-row average LF of 0.33 was able to flush 70% of added salts at McLaren mound + plastic (1.04 kg leached/kg added). In the later years of the Vale, thus showing promising results. On the other hand, this soil has simulation, it reached a value larger than one (1.01) only under the restricted drainage conditions (Singh, 2019), especially in the subsoils, mid-row mounding during 2013–14 and the buried plastic (1.37) which may require different management options for salinity control. during 2012–13. However, LEs remained above one in mid-row Overall the mid-row mound covered with plastic (C) was the most ef- mound + plastic (C) during all seasons and reached a maximum value ficient technique for leaching salt from the root zone and attaining of 1.81 kg salts leached/kg added during 2012–13. Therefore, only the LEs > 1, especially at Padthaway. Salt leaching results were extremely mid-row mound + plastic (C) treatment showed continuous depletion variable as influenced by soil texture and heterogeneity. These techni- of salts from the soil at Padthaway. ques should be adopted cautiously depending on the soil, climate, and

The LEs remained larger than one at Loxton irrespective of irrigation water quality at a particular location. The main drawback of the plastic mulch is its cost, its labour-in- 2.0 tensive application, and environmental risk of plastic residues in the

ycneiciffegnihcaeL 2011-12 soil (Silin and Xujian, 2008; Changrong et al., 2014). These issues need 1.5 2012-13 due consideration before implementing the mounding with plastic 2013-14 strategy although biodegradable plastic mulching (Adhikari et al., 1.0 2016; Saglam et al., 2017; Chen et al., 2019, 2020) or sprayable biodegradable polymer membrane (Filipović et al., 2020) are viable op- 0.5 tions for addressing the environmental issues. Other options of en- 0.0 vironmentally friendly mulching, such as compost, wood chips, and ABCDABCDABCD plant residues, are also available. The application of organic mulches is equally labour intensive, but they have numerous other beneficial im- McL Pad Lox pacts such as increasing soil water retention (Granatstein and Mullinix, Fig. 7. Average salt leaching efficiency in terms of percent of annual salts 2008; Yu et al., 2018), and enhancing soil organic matter, nutrient leached from the total added (kg/kg salts added) and percent of salts leached cycling, and N availability (Neilsen et al., 2003; Sanchez et al., 2003; per m3 of drainage (Dr) (kg/m3 of drainage) at McLaren Vale (McL), Padthaway Hoagland et al., 2008; TerAvest et al., 2010). On the other hand, they (Pad), and Loxton (Lox) under different treatments (A - control, B - Mid-row also have adverse impacts, such as causing tree root diseases, rodent (MR) mound, C - MR mound + plastic, D - Buried plastic). infestation (Merwin and Stiles, 1994), and weed seed distribution

10 V. Phogat, et al. Journal of Hydrology 587 (2020) 125000

(Rowley et al., 2011). Moreover, their role in redirecting rainfall, salts Declaration of Competing Interest dynamics, and leaching seems limited and not yet fully explored. Therefore, the use of plastic mulches still seems to be a relatively better The authors declare that they have no known competing financial alternative if the overall goal is to reduce the soil salinization and interests or personal relationships that could have appeared to influ- leaching of soluble salts from the crop root zone. ence the work reported in this paper.

Acknowledgements 4. Conclusions The authors are thankful to the Australian Water Recycling Centre Leaching of soluble salts from an irrigated soil root zone is necessary of Excellence (Grant ID: AWRCOE 3145) and the Goyder Institute for for sustainable crop production since all water additions, and sub- Water Research (Grant ID: I.1.3) for funding assistance. sequent evaporation and transpiration, will increase salt concentrations. This investigation was carried out to understand the impact of Appendix A. Supplementary data different mid-row management techniques such as mounding and plastic mulching for directing rainfall for reducing salt accumulation in Supplementary data to this article can be found online at https:// the root zone of drip-irrigated grapevines. Four different treatments [no doi.org/10.1016/j.jhydrol.2020.125000. mounding or control (A), mid-row mounding (B), mid-row mounding covered with plastic (C), and plastic buried in the soil in the mid-row References (D)] were evaluated using a numerical model (HYDRUS-2D) for simulating the water balance, root zone salinity dynamics (ECe), and salt Abd El-Mageed, T.A., Semida, W.A., Abd El-Wahed, M.H., 2016. Effect of mulching on dynamics in the soil for multiple seasons (2011–2014) in three grape plant water status, soil salinity and yield of squash under summer-fall deficit irri- growing regions. These vines had been irrigated with varied water gation in salt affected soil. Agric. Water Manage. 173, 1–12. ABS, 2015. Vineyards estimates, Australia, cat. no. 1329, Australian Bureau of Statistics, qualities, ranging from river water (0.2–0.5 dS/m) at Loxton, recycled Canberra. water (1.0–1.3 dS/m) at McLaren Vale, and groundwater (1.9–2.3 dS/ Adhikari, R., Bristow, K.L., Caseya, P.S., Freischmidt, G., Hornbuckle, J.W., Adhikari, B., m) at Padthaway. 2016. Preformed and sprayable polymeric mulch film to improve agricultural water use efficiency. Agric. Water Manage. 169, 1–13. Results demonstrated that mid-row mounding with plastic mulching Allen, R.G., Pereira, L.S., 2009. Estimating crop coefficients from fraction of groundcover (C) was effective at controlling salinity under drip-irrigated grapevines. and height. Irrig. Sci. 28 (1), 17–34. It showed a threefold increase in drainage flux (Dr), a drastic reduction Allen, R.G., Pereira, L.S., Raes, D., Smith, M., 1998. Crop evapotranspiration guidelines in evaporation (E ) (48–54%), and an increase in transpiration (T ) of for computing crop water requirements. FAO Irrigation and Drainage Paper No. 56, s p FAO, Rome, Italy. grapevine (1–16%), as compared to the control (A). Despite such re- Ayers, R.S., Westcot, D.W., 1985. Water Quality for Agriculture. FAO Irrigation and markable impacts of mid-row mounding coupled with plastic mulch, Drainage Paper 29, Food and Agriculture Organization of the United Nations, Rome, soil texture, and the quantity and quality of irrigation water have p. 174. Beven, K.J., Henderson, D., Reeves, A.D., 1993. Dispersion parameters for undisturbed played an important role in controlling salt transport in the soil. partially saturated soil. J. Hydrol. 143, 19–43. Modelling simulations confirmed that in heavy textured highly het- Bezborodov, G.A., Shadmanov, D.K., Mirhashimov, R.T., Yuldashev, T., Qureshi, A.S., erogeneous soils (McLaren Vale site) in vineyards irrigated with re- Noble, A.D., Qadir, M., 2010. Mulching and water quality effects on soil salinity and sodicity dynamics and cotton productivity in Central Asia. Agric. Ecosyst. Environ. cycled water (ECiw = 1.2 dS/m), this treatment, in spite of one and half 138, 95–102. times higher salts leaching than in the other treatments, could not Cao, D., Li, Y.Y., Liu, B.H., Kong, F.J., Tran, L.S.P., 2018. Adaptive mechanisms of soy- bean grown on salt-affected soils. Land Degrad. Dev. 29, 1054–1064. maintain a salt removal efficiency (LEs) > 1 over the simulation Changrong, Y., Wenqing, H., Turner, N.C., Enke, L., Qin, L., Shuang, L., 2014. Plastic-film period. Therefore, the final average profile ECe in the under-vine (UV) mulch in Chinese agriculture: Importance and problems. World Agric. 4 (2), 32–36. region increased from 1.1 to 2.4 dS/m, which is more than the salinity Chen, L., Feng, Q., Li, F., Li, C., 2015. Simulation of soil water and salt transfer under tolerance threshold (2.2 dS/m) for grapevines. Groundwater-irrigated mulched furrow irrigation with saline water. Geoderma 241–242, 87–96. Chen, N., Li, X., Šimůnek, J., Shi, H., Ding, Z., Peng, Z., 2019. Evaluating the effects of (ECiw = 2.2–2.35 dS/m) grapevines on medium-textured soils biodegradable film mulching on soil water dynamics in a drip-irrigated field. Agric. (Padthaway site) with double the irrigation amount compared to Water Manage. 226, 105788. https://doi.org/10.1016/j.agwat.2019.105788. McLaren Vale, and a similar annual rainfall maintained LEs > 1 during Chen, N., Li, X., Šimůnek, J., Shi, H., Shi, J., Ding, Z., Zhang, Y., 2020. The effects of all three seasons, reduced the average EC in the UV region from 5.2 to biodegradable and plastic film mulching on nitrogen uptake, leaching in a drip-irri- e gated sandy field. Agric., Ecosyst. and Environ. 292, 106817. https://doi.org/10. 3.2 dS/m. A leaching fraction of 0.18 was found to leach all annually 1016/j.agee.2020.106817. added salts out of the root zone in the latter soils. Chen, S., Zhang, Z., Wang, Z., Guo, X., Liu, M., Hamoud, Y.A., Zheng, J., Qiu, R., 2016. These results suggest that rainfall redirection techniques have the Effects of uneven vertical distribution of soil salinity under a buried straw layeron the growth, fruit yield, and fruit quality of tomato plants. Sci Hortic. 203, 131–142. potential to play an important role in reducing root zone salts. Still, the Christen, E., DeLange, S., Patti, T. Hornbuckle, J., 2007. Soil salinity in drip irrigated extent of the reduction will vary depending on the soil and agro-cli- vineyards of the MIA. IREC Farmers’ Newsletter, No. 176, Spring 2007, 54-57. matic zone. The requirement for salt removal from the soil is influenced Cleugh, H., 2006. Water and salt balances in Padthaway wine region. Final report of the Project No. RD 04/02-1. to Grape and Wine Research and Development. Corporation. by soil texture, soil heterogeneity, irrigation quantity and quality, Cote, C.M., Bristow, K.L., Charlesworth, P.B., Cook, F.J., Thorburn, P.J., 2003. Analysis of rainfall volume, distribution and intensity, and the salinity tolerance soil wetting and solute transport in subsurface trickle irrigation. Irrig. Sci. 22, threshold of the cropland, thus requires site-specific modelling and 143–156. Cresswell, R.G., Dighton, J., Leaney, F., Vleeshouwer, J., Morrow, D., Harris, M., Stenson, management. M., 2010. Australia-wide Network to Measure Rainfall Chemistry and Isotopic Composition −Final Report of Project MD311. CSIRO, Water for a Healthy Country National Research Flagship, Australia. da Silva, J.R., Rodrigues, W.R., Ferreira, L.S., Bernado, W.P., Paixão, J.S., Patterson, A.E., CRediT authorship contribution statement Ruas, K.F., Viana, L.H., de Sousa, E.F., Bressan-Smith, R.E., Poni, S., Griffin, K.L., Campostrini, E., 2018. Deficit irrigation and transparent plastic covers can save water V. Phogat: Methodology, Software, Data curation, Formal analysis, and improve grapevine cultivation in the tropics. Agric. Water Manage. 202, 66–80. Writing - original draft. T. Pitt: Investigation, Methodology, Project DeGaris, K.A., Walker, R.R., Loveys, B.R., Tyerman, S.D., 2015. Impact of deficit irrigation strategies in a saline environment on Shiraz yield, physiology, water use and tissue administration. R.M. Stevens: Conceptualization. J.W. Cox: ion concentration. Aust. J. Grape Wine Res. 21, 468–478. Supervision. J. Šimůnek: Writing - review & editing. P.R. Petrie: Dong, H., Li, W., Xin, C., Tang, W., Zhang, D., 2010. Late planting of short-season cotton Writing - review & editing. in saline fields of the Yellow River Delta. Crop Sci. 50, 292–300. Dudley, L.M., Ben-Gal, A., Shani, U., 2008. Influence of plant, soil and water properties on the leaching fraction. Vadose Zone J. 7, 420–425. FAO, 1994. Water quality for agriculture. FAO Irrigation and drainage paper. 29 Rev.1.

11 V. Phogat, et al. Journal of Hydrology 587 (2020) 125000

Rome. Qin, S., Zhang, J., Dai, H., Wang, D., Li, D., 2014. Effect of ridge–furrow and plastic- FAO, 2016. AQUASTAT website. Food and Agriculture Organisation of the United Nations mulching planting patterns on yield formation and water movement of potato in a (FAO). http://www.fao.org/nr/water/aquastat/didyouknow/index3.stm. Accessed semi-arid area Shuhao. Agric. Water Manage. 131, 87–94. on 14th September 2019. Ramakrishna, A., Tam, H.M., Wani, S.P., Long, T.D., 2006. Effect of mulch on soil tem- Filipović, V., Bristow, K.L., Filipović, L., Wang, Y., Sintim, H.Y., Flury, M., Šimůnek, J., perature moisture, weeds infestation and yield of groundnut in northern Vietnam. 2020. Sprayable biodegradable polymer membrane technology for cropping systems: Field Crops Res. 95, 115–125. Challenges and opportunities, Environ. Sci. & Technol., doi: 10.1021/acs.est. Ramos, T.B., Šimůnek, J., Goncalves, M.C., Martins, J.C., Prazeres, A., Castanheira, N.L., 0c00909. Pereira, L.S., 2011. Field evaluation of a multi-component solute transport model in Flexas, J., 2016. Genetic improvement of leaf photosynthesis and intrinsic water use ef- soils irrigated with saline waters. J. Hydrol. 407, 129–144. ficiency in C3 plants: why so much little success? Plant Sci. 9407,1–7. Ramos, T.B., Šimůnek, J., Goncalves, M.C., Martins, J.C., Prazeres, A., Pereira, L.S., 2012. Gao, Y., Wu, P.T., Zhao, X.N., Wang, Z.K., 2014. Growth, yield, and nitrogen use in the Two-dimensional modelling of water and nitrogen fate from sweet sorghum irrigated wheat/maize intercropping system in an arid region of north-western China. Field with fresh and blended saline waters. Agric. Water Manage. 111, 87–104. Crops Res. 167, 19–30. Rowley, M.A., Ransom, C.V., Reeve, J.R., Black, B.L., 2011. Mulch and organic herbicide Granatstein, D., Mullinix, K., 2008. Mulching options for Northwest organic and con- combinations for in-row orchard weed suppression. Int. J. Fruit Sci. 11, 316–331. ventional orchards. HortSci. 43, 45–50. Saglam, M., Sintim, H.Y., Bary, A.Y., Miles, C.A., Ghimire, S., Inglis, D.A., Flury, M., 2017. Hanson, B.R., Hopmans, J.W., Šimůnek, J., 2008. Leaching with subsurface drip irrigation Modelling the effect of biodegradable paper and plastic mulch on soil moisture dy- under saline shallow groundwater conditions. Vadose Zone J. 7, 810–818. namics. Agric. Water Manage. 193, 240–250. Hoagland, L., Carpenter-Boggs, L., Granatstein, D., Mazzola, M., Smith, J., Peryea, F., Sanchez, J.E., Edson, C.E., Bird, G.W., Whalon, M.E., Wilson, T.C., Harwood, R.R., Reganold, J.P., 2008. Orchard floor management effects on nitrogen fertility andsoil Kizilkaya, K., Klein, W., Middleton, A., Loudon, T.L., 2003. Orchard floor and ni- biological activity in a newly established organic apple orchard. Biol. Fertil. Soils 45, trogen management influences soil and water quality and tart cherry yields. J.Am. 11–18. Soc. Hort. Sci. 128, 277–284. IPCC, 2014. Intergovernmental Panel on Climate Change. Climate Change 2014: Impacts, Selim, T., Berndtsson, R., Persson, M., 2013. Simulation of soil water and salinity dis- Adaptation, and Vulnerability. http://www.ipcc.ch/report/ar5/wg2. (Accessed on 04 tribution under surface drip irrigation. Irrig. Drain. 62, 352–362. December 2019). Šimůnek, J., van Genuchten, M.Th., Šejna, M., 2016. Recent developments and applica- Jeffrey, S.J., Carter, J.O., Moodie, K.B., Beswick, A.R., 2001. Using spatial interpolation to tions of the HYDRUS computer software packages. Vadose Zone J. 15 (7), 1–25. construct a comprehensive archive of Australian climate data. Environ. Modell. https://doi.org/10.2136/vzj2016.04.0033. Software 16, 309–330. Silin, C., Xujian, W., 2008. The research status of plastic film residual pollution control Klute, A., 1986. Methods of soil analysis. Part 1. Physical and mineralogical methods. and the patent strategy. Agric. Machinery 20, 77–78. 146, 413–423. Singh, A., 2019. Environmental problems of salinization and poor drainage in irrigated Li, X., Shi, H., Simunek, J., Gong, X., Peng, Z., 2015. Modelling soil water dynamics in a areas: Management through the mathematical models. J. Clean Prod. 206, 572–579. drip-irrigated intercropping field under plastic mulch. Irrig. Sci. 33, 289–302. Stevens, R.M., Pitt, T.R., Dyson, C., 2013. Changes in vineyard floor management reduce Li, X., Šimůnek, J., Shi, H., Yan, J., Peng, Z., Gong, X., 2017. Spatial distribution of soil the Na+ and Cl− concentrations in wine grapes grown with saline supplementary water, soil temperature, and plant roots in a drip-irrigated intercropping field with drip irrigation. Agric. Water Manage. 129, 130–137. plastic mulch. European J. Agron. 83, 47–56. Stevens, R.M., Pitt, T.R., Dyson, C., 2012. Managing soil salinity in groundwater irrigated Liu, M., Yang, J., Li, X., Yu, M., Wang, J., 2013. Numerical simulation of soil water dy- vineyards. Final Report to the National Program for Sustainable Irrigation. Project namics in a drip irrigated cotton field under plastic mulch. Pedoshere 25 (5), Number: CIF5121. http://www.npsi.gov.au/products/npsi1212 (Accessed on 26 620–635. November 2019). Merwin, I., Stiles, W., 1994. Orchard groundcover management impacts on apple tree TerAvest, D., Smith, J.L., Carpenter-Boggs, L., Hoagland, L., Granatstein, D., Reganold, growth and yield, and nutrient availability and uptake. J. Am. Soc. Hort. Sci. 119, J.P., 2010. Influence of orchard floor management and compost application timing 209–215. on nitrogen partitioning in apple trees. HortSci. 45, 637–642. Neilsen, G.H., Hogue, E.J., Forge, T., Neilsen, D., 2003. Mulches and biosolids affect van Genuchten, M.Th., 1980. A closed form equation for predicting the hydraulic con- vigor, yield and leaf nutrition of fertigated high density apple. HortSci. 38, 41–45. ductivity of unsaturated soils. Soil Sci. Soc. Am. J. 44, 892–898. Phogat, V., Cox, J.W., Kookana, R.S., Šimůnek, J., Pitt, T., Fleming, N., 2019. Optimizing van Leeuwen, C., Darriet, P., 2016. The impact of climate change on viticulture and wine the riparian zone width near a river for controlling lateral migration of irrigation quality. J. Wine Econ. 11, 150–167. water and solutes. J. Hydrol. 570, 637–646. Wang, Y., Xie, Z., Malhi, S.S., Vera, C.L., Zhang, Y., Wang, J., 2009. Effects of rainfall Phogat, V., Cox, J.W., Šimůnek, J., 2018a. Identifying the future water and salinity risks harvesting and mulching technologies on water use efficiency and crop yield inthe to irrigated viticulture in the Murray-Darling Basin. South Australia. Agric. Water semi-arid Loess Plateau. China. Agric. Water Manage. 96, 374–382. Manage. 201, 107–117. Wang, Z., Jin, M., Šimůnek, J., van Genuchten, M.Th., 2014. Evaluation of mulched drip Phogat, V., Cox, J.W., Šimůnek, J., Hayman, P., 2018b. Modelling water and salinity risks irrigation for cotton in arid Northwest China. Irri. Sci. 32, 15–27. to viticulture under prolonged sustained deficit and saline water irrigation. J. Water Wichelns, D., Qadir, M., 2015. Achieving sustainable irrigation requires effective man- Climate Change. https://iwaponline.com/jwcc/article-pdf/doi/10.2166/wcc.2018. agement of salts, soil salinity, and shallow groundwater. Agric. Water Manage. 157, 186/229889/jwc2018186.pdf. 31–38. Phogat, V., Skewes, M., Cox, J.W., Sanderson, G., Alam, J., Šimůnek J., 2014. Seasonal Yu, Y.Y., Turner, N.C., Gong, Y.H., Li, F.M., Fang, C., Ge, L.J., 2018. Benefits and lim- simulation of water, salinity, chloride and nitrate dynamics under drip irrigated itations to straw- and plastic-film mulch on maize yield and water use efficiency: A mandarin (Citrus reticulate) and assessing management options for drainage and meta-analysis across hydrothermal gradients. European J. Agron. 99, 138–147. nitrate leaching. J. Hydrol. 513, 504–516. Zhang, X., Walker, R.R., Stevens, R.M., Prior, L.D., 2002. Yield-salinity relationships of Phogat, V., Skewes, M. A., McCarthy, M. G., Cox, J. W., Šimůnek, J., Petrie, P. R., 2017. different grapevine (Vitis Vinifera L.) scion-rootstock combinations. Aust. J. Grape Evaluation of crop coefficients, water productivity, and water balance components Wine Res. 8, 150–156. for wine grapes irrigated at different deficit levels by a sub-surface drip. Agric. Water Zhang, Y.L., Feng, S.Y., Wang, F.X., Binley, A., 2018. Simulation of soil water flow and Manage. 180, 22–34. heat transport in drip irrigated potato field with raised beds and full plastic-film Pitt, T., Cox, J., Phogat, V., Fleming, N., Grant, C., 2015. Methods to increase the use of mulch in a semiarid area. Agric. Water Manage. 209, 178–187. recycled wastewater in irrigation by overcoming the constraint of soil salinity. Zhao, Y., Pang, H., Wang, J., Huo, L., Li, Y., 2014. Effects of straw mulch and buried straw Australian Water Recycling Centre of Excellence, Brisbane Australia http:// on soil moisture and salinity in relation to sunflower growth and yield. Field Crops www.australianwaterrecycling.com.au/_literature_148442/ Res. 161, 16–25. Increased_use_of_water_recycling_in_irrigation_-_Final_Report. Zhao, Y., Zhai, X.F., Wang, Z.H., Li, H.J., Jiang, R., Hill, R.L., Si, B., Feng, H., 2018. Qadir, M., Quillerou, E., Nangia, V., Murtaza, G., Singh, M., Thomas, R.J., Drechsel, P., Simulation of soil water and heat flow in ridge cultivation with plastic film mulching Noble, A.D., 2014. Economics of salt-induced land degradation and restoration. Nat. system on the Chinese Loess Plateau. Agric. Water Manage. 202, 99–112. Resour. Forum 38, 282–295.