通过增加渗透和减少蒸发，覆盖可以改善半干旱地区的土壤水环境和果园的苹果产量

A R T I C L E I N F O

Handling Editor - Dr. B.E. Clothier

Keywords: Mulching Infiltration Evaporation Soil water Orchards Yield

A B S T R A C T

Infiltration and evaporation are two important components of soil water balance in semiarid areas. For local orchards using drip irrigation, rainfall and irrigation events have a profound impact on soil infiltration and evaporation. As an effective water conservation technique, soil surface mulching is widely used in semiarid areas. The aim of this study was to investigate the effect of mulching (e.g., horticultural fabric mulching and corn straw mulching) on soil infiltration and evaporation after irrigation and rainfall. Through field experiments, we investigated soil infiltration and evaporation during three wet-dry cycles and for a rainy month and studied the effects of mulching on soil physical properties and apple yields. The results showed that both horticultural fabric mulching (FM) and straw mulching (SM) improved soil physical properties by reducing soil bulk density and increasing porosity and that SM had a better effect. Mulching increased infiltration after irrigation and heavy rain, while SM reduced infiltration when light rain occurred. Regardless of whether the soil was in the energy- limited stage or falling-rate stage, mulching can effectively reduce soil evaporation. We determined that the threshold of soil water content (SWC) between the energy-limited stage and falling-rate stage was 22.09–22.75%. When SWC is lower than the threshold, there is a positive significant correlation between evaporation and SWC, while no significant correlation was found when it is above the threshold. Moreover, mulching effectively increased apple yields and irrigation water use efficiency (IWUE) for three consecutive years. Therefore, both FM and SM have potential for application in orchards in semiarid areas to improve soil and moisture conditions in the root zone.

1. Introduction

Apple trees are one of the most important perennial fruit trees planted in the Loess Plateau of China. However, due to drought and serious soil erosion, sustainable development of the apple industry in this region has been seriously affected (Zhong et al., 2019). In recent years, the apple industry in Shaanxi Province has expanded rapidly from semi-humid areas to semi-arid areas, which further aggravates the contradiction between supply and demand of water resources (Yang et al., 2020). Therefore, maintaining sufficient soil moisture storage is very important for this industry. In local apple orchards, rainfall is the main source of soil water supplements. However, due to the character- istics of scarce and uneven rainfall, drip irrigation was introduced to supplement soil moisture. Evaporation and infiltration are two types of soil water movement for maintaining soil water balance. In the context of climate change, increasing infiltration from precipitation and irriga- tion and decreasing soil evaporation are therefore of great importance to

the sustainability of agriculture in water-limited regions (Li et al., 2018; Cao et al.,2009).

Water infiltration is an important process of mutual transformation of rainwater, irrigation water, surface water, soil water and ground- water. Water infiltration is affected by irrigation amounts, precipitation characteristics, canopy interception capacities and soil hydraulic char- acteristics (Khan et al., 2016). In general, when rainfall occurs, due to the low initial soil water content (SWC), a large amount of rainwater infiltrates into the soil and is converted into soil water. If the rainfall amount is large, as infiltration progresses, soil water gradually becomes saturated, and the infiltration rate gradually becomes lower than the rainfall intensity, rainwater gathers on the soil surface to form surface water and runoff and is ultimately lost. Furthermore, surface runoff causes severe soil degradation and loss of sustainable production through soil erosion and nutrient loss (Liu et al., 2012, 2018; Prosdocimi et al., 2016).

Under the influence of atmospheric radiant heat, soil water

* Correspondence to: No. 26, Weihui Road, Yangling, Shaanxi 712100, China.

E-mail addresses: ly0225@126.com (Y. Liao), nschx225@nwafu.edu.cn (H.-X. Cao).

https://doi.org/10.1016/j.agwat.2021.106936

Received 9 December 2020; Received in revised form 18 April 2021; Accepted 19 April 2021

Available online 29 April 2021

0378-3774/© 2021 Elsevier B.V. All rights reserved.

evaporates at the soil-atmosphere interface. However, this portion of soil water loss is often regarded as invalid water consumption because it does not participate in biomass formation or yields (Chen et al., 2015; Ram et al., 2013). Some scholars have found that when evaporation decreases due to the controlling effect of mulching, plant transpiration increases in some field crops (Singh et al., 2011; Li et al., 2008). In field environments, soil water evaporation can be divided into two stages. When rainfall or irrigation have just ended, SWC is high, and soil evaporation has nothing to do with SWC and is only determined by meteorological factors (e.g., air temperature, solar radiation, wind speed and air humidity), this stage is called “energy-limited stage”. During this energy-limited stage, the soil evaporation rate is relatively constant and occurs at a maximum rate that is limited only by meteorological con- ditions (Zribi et al., 2015). When SWC drops to a certain threshold, the “energy-limited stage” ends. Due to the low SWC, the ability of soil to transport water upward is weakened, which makes it difficult to meet the demand of atmospheric evaporation in the surface layer. The evaporation rate depends on the demand for atmospheric evaporation and the ability of soil to transport water upward. During this stage, the evaporation rate is reduced in proportion to the water available at the soil surface, and this stage is called “falling-rate stage”(Allen et al., 1998).

As an effective water conservation technique, soil surface mulching is widely used in arid and semiarid areas (Chakraborty et al., 2008; Zhang et al., 2009; Zribi et al., 2015). Mulch forms a physical isolation layer on the soil surface, which slows down and hinders exchange of water and energy between the soil and atmosphere. The effectiveness of mulching for soil evaporation control and increasing infiltration has been documented in some studies (Li et al., 2018; Khan et al., 2016; Wang et al., 2015; Wang et al., 2009). In addition to the benefits of water conservation, previous studies have shown that soil surface mulching was also reported to regulate soil temperature variations (Olasantan, 1999; Awe et al., 2015; Sarkar et al., 2007), alter soil microbiology (Steinmetz et al., 2016; Carney and Matson, 2005), improve soil physical properties and soil fertility (Smets et al., 2008; Chen et al., 2014; Rodrigues et al., 2013; Jord´an et al., 2010), and affect crop growth, physiology and, ultimately, yields and water use efficiency (Zhang et al., 2014; Wang et al., 2015; Abd El-Wahed and Ali, 2013).

In our previous study (Liao et al., 2021), we studied the effect of mulching on SWC during the whole growth period from 2018 to 2019, and found that mulching effectively increased SWC, but the reason was not clear. Infiltration after rainfall and irrigation is the main water replenishment method in this area, while evaporation is the main loss method of field soil water. We hypothesize that given these processes, mulching exerts a great impact on the Soil-Plant-Atmosphere Contin- uum (SPAC) and is different after irrigation and rainfall. This is because water can enter the soil directly during irrigation, while it must pass through the mulching layer first before reaching the soil during rainfall. Regardless of whether rain or irrigation has occurred, mulching can significantly reduce evaporation because it hinders water and vapor exchange between the soil and atmosphere and reduces the amount of heat radiation reaching the ground. However, the effects of mulching on soil infiltration are different after irrigation and rainfall because mulching changes the structure of the underlying surface of the field, and the mulching itself absorbs water, which depends on the mulching material.

Several field experiments were conducted on the control of soil evaporation and regulation of infiltration under mulching, respectively. However, there is a lack of mechanistic research on the effects of mulching on soil evaporation and infiltration after rainfall and irriga- tion, especially for orchards in semiarid areas. During the fruit expan- sion stage (late June to late August), the SWC changed dramatically. In the early fruit expansion stage, the SWC was at the lowest level throughout the growth season due to higher evapotranspiration and less rainfall, while in the late fruit expansion stage, a large amount of rainfall infiltration will lead to the rapid increase of SWC (Liao et al., 2021).

Both soil evaporation and infiltration at this stage are representative. Therefore, we chose the fruit expansion stage to study the effect of Mulching on soil evaporation and infiltration. The specific objectives were to (1) study the infiltration and evaporation processes after rainfall and irrigation and the effects of mulching on them; (2) analyse the relationship between soil evaporation and soil water content; and (3) investigate the effects of mulching on yields and irrigation water use efficiency (IWUE).

2. Material and methods

2.1. Experiment location

The study was carried out from 2018 to 2020 at a modern agriculture apple demonstration orchard located in Qingshuigou Village, Zizhou County, Yulin City, Shaanxi Province, northwest China (37◦27′ N, 110◦2′ E, altitude: 1020 m). The experimental site is in a typical hilly gully region of the Loess Plateau and has a continental temperate climate with an annual average sunshine duration of over 2543.3 h. The annual average temperature is 9.7 ◦C, and the annual average frost-free period is 170 days. The average annual rainfall is 453 mm, and more than 60% occurs from July to September. The test soil is classified as sandy loam with favourable hydraulic properties (Gao et al., 2018). The mean pH and field capacity (θf) of the one-metre soil layer are 8.5 and 0.30

cm3⋅cm—3, respectively. An automated weather station was installed in

the orchard to monitor meteorological factors such as precipitation, wind velocity, wind direction, solar radiation, atmospheric pressure, air temperature and relative air humidity.

2.2. Experimental design

Eight-year-old ‘Honeycrisp’ apple trees (Malus pumila Mill) with similar trunk diameters and bark depths were selected as the experi- mental trees. The apple trees were planted in an east-west orientation

with a row spacing of 3.5 m and plant spacing of 2 m (1429 plants ha—1).

Each treatment included three rows of apple trees. Three apple trees in each treatment with similar growth vigour and canopy were selected as the experimental trees.

Four treatments were employed: horticultural fabric mulching with drip irrigation (FM), corn straw mulching with drip irrigation (SM), clear tillage with drip irrigation (TL), and clear tillage with rain feeding (CK). The randomized complete block design with a split-plot arrange- ment were adopted in the experiment, each treatment was repeated three times (Fig. 1). More specific details of the mulching treatments are provided in our other report (Liao et al., 2021). In April every year, TL and CK were cultivated to a tillage depth of 30 cm by a mechanical cultivator, while no tillage was implemented for SM and FM. Other agronomic measures were consistent, such as fertilization, weeding, pruning and insecticide spraying.

2.3. Irrigation and soil water monitoring

According to local soil water consumption rates, vertical SWC pro- files were monitored for an approximately 10 days interval with a TRIME-T3 tubular TDR (IMKO Ltd, DE) at 20 cm intervals over the 80 cm-depth soil layer. Two trim tubes for each treatment were installed at horizontal distances of 40 and 80 cm from the tree row (Fig. 1). The SWC for each treatment was the average of the SWCs that were measured by the two trim tubes. Different mulching treatments received the same irrigation amounts, and the irrigation amount applied was defined based on the volumetric SWC of TL. When the SWC of TL was lower than the lower irrigation limit (70% θf), irrigation was carried out. More specific details on irrigation systems and irrigation events are given in Liao et al. (2021).

From June 27 to the end of August 2020, nearly daily continuous monitoring of SWC was carried out to observe SWC variations after rain

Fig. 1. Layout of the experiment.

and irrigation and to investigate their relationships with evaporation.

Soil water storage (SWS) is a key component of soil water balance and is defined by the formula (1):

SWS = θh (1)

Here, θ is the mean volumetric soil water content of a specific soil layer (%), it should be noted that in the rainy August, some measurements of SWC are not carried out on the day of rain or the next day due to the surface water and mud in the field (Table 4); h is the depth of a specific soil layer (mm).

Increased SWS (ΔSWS) after rain or irrigation can be used to assess infiltration efficiency (Song et al., 2020) and was calculated using the following formula (2):

ΔSWS = SWS1 — SWS0 (2)

Here, SWS1 is soil water storage after rain or irrigation and SWS0 is soil water storage before rain or irrigation.

2.4. Measurements of soil bulk density and porosity

After the apple harvests in 2019 and 2020, soil bulk density, porosity, capillary porosity and non-capillary porosity were measured at 20 cm intervals over the 60 cm-depth soil layer using the cutting ring method (Bao, 2007). The soil bulk density and porosity in 2018 were not measured because the orchard ground was just covered in April 2018, which may have little impact on the soil.

2.5. Monitoring of evaporation

Soil evaporation was measured by weighing microlysimeters (MLs) at 8:00 every day during fruit expansion (late June to late August) in 2020. Each microlysimeter consisted of an inner cylinder made of PVC pipe and an outer cylinder made of iron sheet. The inner cylinder had a diameter of 10.5 cm and height of 19.5 cm, and the outer cylinder had a slightly larger size to facilitate the tight installation of the inner cylinder into the outer cylinder. A handle was attached to the inner cylinder to facilitate its periodic extraction from the outer cylinder.

Two MLs each treatment were installed on both sides of the tree row, with a horizontal distance of 60 cm from the tree row of each subplot. First, on the day after a rain or irrigation event, the inner cylinder was inserted into the soil to extract the undisturbed soil, and this cylinder

was then was packed with gauze at the bottom. Then, it was weighed by a portable balance (0.01 g precision) and was placed back into the outer cylinder. After that, the inner cylinder was weighed at 8:00 each day to measure the amount of evaporation of the previous day. Long et al. (2010) found that if the soil was not replaced within 15 days, the ac- curacy of soil evaporation was still high. Therefore, the MLs were renewed with undisturbed soil after the next rain (greater than 10 mm) or irrigation or use for more than 15 days. Moreover, for the mulching treatments, the MLs were covered by the mulching layer. In the SM treatment, a 20 cm by 20 cm metal screen was placed on the top of the MLs to prevent straw debris from falling into the MLs. This process was followed during three wet-dry cycles: 27 June (irrigation 15.3 mm) – 7 July, 9 July (one day after rainfall 17.5 mm) – 19 July and 20 July (one day after irrigation 17.2 mm) to 29 July. Due to the high frequency-high magnitude rainfall in August, it was difficult to measure SWC and soil evaporation when rainfall amounts were large. Therefore, in August, there were no long-term continuous observations of soil moisture con- tent and soil evaporation.

In order to eliminate the influence of climate conditions and canopy shading on soil evaporation, we introduced the relative evaporation rate (ER) under the trees, which was calculated using the following formula (3):

ER = E/Epan (3)

Here, E is soil evaporation measured by MLs and Epan is the water surface evaporation measured by two standard 20 cm diameter evaporation pans that were installed on both sides of the tree row at 60 cm horizontal distances from the tree rows. More details on measurements of water surface evaporation are given in Zhao et al. (2020).

2.6. Leaf area index, fruit yield and IWUE

The leaf area index (LAI) was measured by an LI-2200C canopy analysis system (LI-COR, Nebraska, USA). To reduce the influence of strong light, measurements of LAI were performed before sunrise or after sunset.

Apples were harvested and the yield per treatment were weighed on 17 September 2018, 21 September 2019 and 10 September 2020. Irri- gation water use efficiency (IWUE) was calculated using the following formula (Du et al. 2017) (4):

IWUE = 1/10 × (Y — Y1)/I (4)

Here, Y is the yield under irrigation (kg ha—1), Y1 is the yield under rainfed conditions, and I is the irrigation amount (mm).

2.7.
Statistical analysis

Analysis of variance (ANOVA) was performed with SPSS 25.0 Sta- tistics (SPSS Inc., Chicago, IL, USA). The least significant difference

Non-capillary porosity

50

Capillary porosity

1.50

Soil bulk density

50

1.50

40 1.25 40

1.25

30

0

CK TL

FM SM

1.00

30

0

CK TL

FM SM

1.00

a. 0-20cm (2019)

50

1.50 50

d. 0-20cm (2020)

1.50

40 1.25 40

1.25

30

0

CK TL

FM SM

30

1.00 0

CK TL

FM SM

1.00

b. 20-40cm (2019)

50

1.50 50

e. 20-40cm (2020)

1.50

40 1.25 40

1.25

30

0

CK TL

FM SM

1.00

30

0

CK TL

FM SM

1.00

c. 40-60cm (2019) f. 40-60cm (2020)

Fig. 2. Soil bulk density and porosity in 0–60 cm soil layer in 2019 and 2020. The error bars represent the standard deviations (black error bars represent the standard deviations of porosity). Different letters within a column indicate significant differences among treatments at P = 0.05 level. TL: clear tillage with drip irrigation; FM: horticultural fabric mulching with drip irrigation; SM: corn straw mulching with drip irrigation; CK: clear tillage with rain feeding.

(LSD) at a 5% significance level was used to determine any significant differences. Images were generated using Origin 2018 (OriginLab Inc., Massachusetts, USA).

3. Results

3.1. Soil bulk density and porosity

In the second (2019) and third (2020) years of the experiment, the soil bulk density of the 0–40 cm soil layer was significantly reduced by mulching, especially in the 0–20 cm soil layer, and there were no sig- nificant differences between the two years (Fig. 2). With increasing soil depth, the soil bulk density first increased and then decreased, with a minimum at 0–20 cm and maximum at 20–40 cm. In the 0–20 cm soil layer, the mean soil bulk densities of FM and SM in the two years were

1.32 g/cm3 and 1.28 g/cm3, which were 3.1% and 6.4% lower than that

of TL (1.36 g/cm3), respectively. In the 20–40 cm soil layer, the mean soil bulk densities of FM and SM in the two years were 1.44 g/cm3 and

1.45 g/cm3, which were 2.7% and 2.5% lower than that of TL (1.48 g/ cm3), respectively. The soil bulk density in the 0–40 cm layer of SM was

significantly lower than that of FM except in the 20–40 cm soil layer in 2020.

In the 0–40 cm soil layer, capillary porosity, non-capillary porosity and total porosity were increased significantly by mulching in the two years (Fig. 2). In contrast to soil bulk density, with increasing soil depth, the porosity first decreased and then increased, with a minimum at 20–40 cm and maximum at 0–20 cm. In the 0–20 cm soil layer, the mean soil porosities of FM and SM in the two years were 43.17% and 45.83%, respectively, which were 4.5% and 10.9% higher than that of TL (41.33%). In the 20–40 cm soil layer, the mean soil porosities of FM and SM of the two years were 39.6% and 40.4%, which were 5.0% and 7.0% higher than that of TL (37.72%), respectively. In the 0–20 cm soil layer, SM soil porosity was significantly higher than FM soil porosity; however, there was no significant difference in the 20–40 cm soil layer.

Regardless of the soil bulk density or porosity, no significant differ- ence was observed in the 40–60 cm soil layer under the different tillage practices. There were no significant differences in porosity or soil bulk density under irrigation in the two years, and no significant difference was observed between TL and CK.

3.2. Variations in soil water after irrigation and rain

The meteorological factors in the three wet-dry cycles were generally the same (Table 1). Fig. 3 shows the variations in SWC of the three tillage practices at different soil depths after rain or irrigation. After irrigation, the SWC of the 0–40 cm soil layer increased rapidly and then gradually returned to a lower level. The increase in SWC in the 0–20 cm soil layer was greater than that in the 20–40 cm soil layer, which was related to water replenishment in the upper soil that occurred first after rainfall or irrigation. In the SM and FM treatments, the SWC of the 40–60 cm soil layer also increased, but there was no similar increase in the TL treat- ment. The SWC of the 60–80 cm soil layer did not change after irrigation and showed a gradually decreasing trend in July. When irrigation

Table 1

Meteorological conditions of three wet-dry cycles in fruit expansion stage in 2020.

occurred, we found that the increase in soil water storage (SWS) was higher than the irrigation amount (Table 2), which could be due to accumulation of water around the emitters after irrigation. In the FM and SM treatments, the increased SWS was higher than that in the TL treatment Table 3.

The increase in SWS was always lower than rainfall due to canopy interception, surface runoff and soil evaporation. After the rain on 8 July, the TL treatment (15.04 mm) exhibited the greatest increase in SWS, while the lowest increase occurred in the SM treatment (12.80 mm). In addition, some small rainfall amounts occurred during the three cycles. When the rainfall was greater than 4 mm, the SWC of the 0–20 cm soil layers in TL and FM increased slightly, while SM only increased when larger rainfall occurred (7.8 mm on 25 July) during the three cycles Fig. 4 and Fig. 5.

The high frequency-high magnitude rainfall that occurred in August had a great impact on SWC, especially the continuous rainfall from August 4 to August 6 (Fig. 6 and Table 4). A total of 135.8 mm of rain fell from August 4 to August 6, which led to a sharp increase in SWC, and the increased SWS of SM (91.1 mm) was higher than those of TL (61.3 mm) and FM (78.6 mm). After another large continuous rainfall event

(rainfall of 41.4 mm from 16 Aug to 18 Aug), the same trend was pre- sent, and the increased SWS exhibited a trend of SM (34.9 mm) > FM (31.9 mm) > TL (26.7 mm). However, when the rainfall amounts were

small, SM always exhibited the smallest SWS increment (Table 4). After rainfall in August, the SWC of the 0–20 cm soil layer under the three tillage practices significantly improved, and no significant differences appeared among them (Fig. 6). However, in the deeper soil layer (80 cm), significantly more water recharge occurred in the two mulch- ing treatments.

3.3. Variations of evaporation after irrigation or rain

We analysed the relationship between soil evaporation and SWC in the 0–20 cm soil layer. The relative evaporation rate was introduced to eliminate the influence of meteorological factors and canopy shading on soil evaporation. Fig. 5 shows the relationship between the relative soil evaporation rate and SWC in the three wet-dry cycles. Thresholds of SWC between two evaporation stages were found in the process of soil evaporation. The thresholds for TL, FM and SM were 22.09%, 22.75% and 22.52%, respectively. When the SWC was above the threshold, the relative soil evaporation changed little with SWC, and there was no significant correlation between them. At this time, soil evaporation is in the energy-limited stage. However, when the SWC was below the threshold, the relative soil evaporation rate decreased linearly with the decrease in SWC, and extremely significant correlations were found in both the bare soil and mulching treatments. At this time, soil evapora- tion is in the falling-rate stage. Due to rainfall recharge, the SWC was above the threshold on the date we monitored in August, and no sig- nificant correlation was found between the relative evaporation rate and SWC (Fig. 7). This was consistent with the results in the energy-limited stage in the three wet-dry cycles.

Despite some differences, the same soil evaporation pattern occurred in the three wet-dry cycles. In the early stage of the wet-dry cycle, under

expanding

Irrigation (mm)

Rainfall (mm)

0

28

10

24

20

20

16

6/27 7/2 7/7

28

7/12 7/17

a. TL Date (month/day)

30

7/22 7/27

0

10

24

20

20

16 30

6/27 7/2 7/7 7/12 7/17 7/22 7/27

b. FM Date (month/day)

0

28