Yield and water use efficiency of dryland potato in response to plastic film mulching on the Loess Plateau

ABSTRACT

Water shortage is a key factor limiting potato yield in dryland farming regions, and the practice of soil mulching with plastic film has proven to be effective for increasing crop yield and improving soil and water conservation. This study investigated the effects of plastic mulching on potato yield and water use efficiency (WUE) on the Loess Plateau of China based on a synthesis of 44 publications (1987–2016) from 17 sites. Our results indicated that relative to no mulching, plastic mulching significantly increased the yield and WUE by 29.2% and 31.7%, respectively. The effects on yield under different mulching methods were ranked as follows: ridge-furrow mulching (RFM) > flat mulching (FM), and the effects on yield with different mulching extents were ranked full-area mulching (FAM) > partial-area mulching (PAM). The effects of plastic mulching on the yield using different colors of film were ranked as white plastic mulching (WPM) > black plastic mulching (BPM) and under different rainfall levels as (precipitation < 400 mm) > (precipitation > 400 mm). The rankings of the effects of plastic mulching on WUE were the same as for potato yield. The yield-evapotranspiration (ET) relationship could be described using a quadratic polynomial, and the yield response curve to ET under plastic mulching fell above the no-mulching curve, indicating that plastic mulching can result in higher yields than no mulching when ET is equal. The yield-ET relationship indicated that potatoes required an ET of 334 and 360 mm to achieve the highest yield under plastic mulching and no mulching, respectively, and potato WUE increased linearly with yield under both plastic mulching and no mulching. This study indicated that RFM, FAM and WPM are better potato management measures and that there is potential to improve the yield and WUE on the Loess Plateau.

KEYWORDS:

• Plastic film mulching
• Loess Plateau
• yield
• water use efficiency
• potato
• economic benefits

Introduction

Drylands account for approximately 45% of the Earth’s land area (Schimel 2010). In China, rain-fed agricultural area covers approximately one-third of the arable land, approximately 40% of which is located on the Loess Plateau (Li 2004), and water stress is the main factor limiting agricultural production in this region (Zhang et al. 2009). Due to global climate change, the incidence of drought on the Loess Plateau has dramatically intensified in the past 50 years; the average annual temperature has climbed sharply, and the average annual precipitation has tended to decrease (Xiao et al. 2013). These changes can cause fluctuations in crop yield in arid and semi-arid environments (Lobell et al. 2008), increasing crop water-use efficiency (WUE) in these regions plays an important role in increasing crop yields and promoting the sustainable development of rain-fed agriculture.

On the Loess Plateau, the average annual precipitation ranges from 300 to 600 mm, and more than 90% of farmland cannot be irrigated (Fan, Stewart, Yong, et al. 2005; Wang et al. 2016). In addition, the precipitation is unevenly distributed with more than 60% of the annual rainfall occurring in the summer months (July to September) (Li et al. 2000). Furthermore, most rainfall events occur in the form of low-intensity precipitation that cannot be efficiently utilized by crops, while a few rainfall events are thunderstorms that generally cause tremendous soil erosion (Wang et al. 2008). As one of the main staple food crops in this region, potato has traditionally been planted using a flat or row-drilling method, but such conventional practices have led to considerable loss of soil moisture and poor soil physical conditions (Huang et al. 2003). Due to the scarce and erratic rainfall combined with improper field management in the region, the potato yield is relatively low, and the WUE and economic benefits of potato cultivation are also low (Hou et al. 2010). Thus, enhancing the use of rain water and improving field management to increase potato yield are necessary.

Plastic film mulching first came to China in 1978 and has since been widely used, especially in the northwestern rain-fed regions (Zhao et al. 2010), because this practice can increase crop yield (Zhang et al. 2011). Plastic film can retain soil water content, allow full utilization of limited rainfall (Zhou et al. 2009), increase the temperature of the topsoil (Wang et al. 2005), reduce the loss of fertilizer through eluviation, increase nutrient availability (Haraguchi et al. 2004) and decrease infestation by many annual and perennial weeds (Cohen et al. 2008); therefore, its use may lead to increases in yield and WUE (Li et al. 2013). However, due to prolonged higher soil temperatures (Wang et al. 2009) or higher CO₂ concentrations in the micro-environment that may result from its use (Li, Song, et al. 2004), performing plastic mulching during an entire growing season could greatly reduce the crop yield. In addition, large amounts of plastic film residue may have a detrimental effect on the environment, soil structure, water and nutrient transport and crop growth (Liu et al. 2014).

The relationship between crop yield and water use is a major issue in agricultural research in the arid and semi-arid areas of the world (Musick et al. 1994; Zhang et al. 1999; Fan, Stewart, Payne, et al. 2005). The yield-water use relationship takes different forms and varies with soil texture, climate, crop type and variety, and irrigation, fertilization, and tillage practices. Kang et al. (Kang et al. 2002) reported a curvilinear yield-evapotranspiration (ET) relationship for winter wheat when limited irrigation was applied on the Loess Plateau of China, and similar relationships have been found for wheat under long-term nitrogen fertilization regimes (Zhong and Shangguan 2014) and under five mulching and three tillage practices in northwest China (Wang and Shangguan 2015). Other researchers have showed that grain yield is linearly related to ET (Huang et al. 2004; Fan, Stewart, Payne, et al. 2005). Crop WUE values are primarily related to yield, and some studies have indicated that WUE increases linearly with grain yield (Zhong and Shangguan 2014; Wang and Shangguan 2015) while others have found a quadratic WUE-yield relationship (Musick et al. 1994; Zhang, Wang, You and Liu 1999; Huang et al. 2004). In cereals, such as wheat, the yield-water use relationship has been widely investigated, but the potato yield-water use relationship under plastic mulching remains unclear. Therefore, investigating the potato yield-water use relationship is of great importance to sustaining rainfed agriculture on the Loess Plateau.

Although many field experiments have been conducted to assess the impacts of plastic film mulching on potato yields and WUE, no comprehensive, quantitative analyses of the available published data have been performed thus far. The present study investigated the effects of plastic mulching on the yield, WUE and ET of potatoes through a synthesis of published results. The objectives of this study were (1) to quantify the responses of potato yield, WUE and ET to plastic mulching and establish the potato yield-water use relationship on the Loess Plateau and (2) to compare the effects of different plastic mulching methods, mulching areas, colors of plastic film, and levels of rainfall on yield and WUE.

Materials and methods

Data compilation

A comprehensive literature search of electronic databases, including the Web of Science (http://www.isiknowledge.com/) and the China National Knowledge Infrastructure (http://www.cnki.net/), was conducted. The literature search terms included ‘mulch’ or ‘mulching’, ‘potato’ or ‘Solanum tuberosum’, ‘yield’, ‘water-use efficiency’ or ‘evapotranspiration’ in the article title, abstract, and keywords. A flow diagram of the data search and collection process is presented in Figure 1. Two investigators independently screened the titles, abstracts and full texts of the studies in the search based on the following criteria: (1) field experimental studies must have compared a plastic film mulching treatment against a no-mulching control (conventional tillage), (2) potatoes must not have been irrigated during the entire potato-growing season, and (3) for multifactorial experimental studies, only observations of control and plastic mulching were included (i.e. interactions between treatments were excluded). In addition, the mean values, standard deviations and sample sizes for the yield, WUE, and ET data had to be directly available or calculable from the data obtained during the study. After the search results were carefully checked, 44 studies (6 in English and 38 in Chinese) conducted at 17 sites met the selection criteria (Figure 2). Two investigators independently extracted the relevant data from each study, and if necessary, the third investigator was consulted to resolve any disagreements. If the data used for analysis were in the form of figures or charts, they were transformed from digital versions into numerical values using GetData Graph Digitizer version 2.25 (http://getdata-graph-digitizer.com/).

Figure 1. Flowchart of the process of obtaining data from the literature to build a database for this study.

Figure 2. Map of the Loess Plateau. The locations of the potato field experiments reported in the peerreviewed literature are shown on the map.

Regarding the effects of plastic mulching on potato yield, WUE, and ET, we considered four groups of controlling factors: the mulching method, mulching extent, color of plastic film, and precipitation (see Table 1 for details). The plastic mulching methods were divided into two subgroups, ridge-furrow mulching (RFM) and flat mulching (FM), and the mulching extent was categorized as either full-area mulching (FAM) or partial-area mulching (PAM). There were two types of plastic film color, black plastic mulching (BPM) and white plastic mulching (WPM). Relevant studies were divided into two subgroups according to annual precipitation levels (< 400 mm and > 400 mm), as described in previous reports (Li and Xiao 1992; Huang et al. 2015).

Table 1. Brief descriptions of the materials and methods associated with the different mulching methods, mulching areas and colors of plastic film used in potato field experiments on the Loess Plateau.

Item	Mulching practice	Brief description
Mulching method	Conventional tillage (CT)	The soil was plowed without mulching until the next sowing (no mulching)
	Flat mulching (FM)	Flat planting and mulching with plastic film 7 or 8 μm thick
	Ridge-furrow mulching (RFM)	Ridge-furrow planting with plastic film mulching; the ridge height was 10–30 cm, and the ridge breadth was 40–80 cm
Mulching area	Full area mulching (FAM)	The entire planting area (flat or ridge-furrow) was mulched with plastic film
	Partial area mulching (PAM)	Alternating strips (flat or ridge) mulched with plastic film and bare land
Color of plastic film	Black plastic mulching (BPM)	Mulching with black plastic film 7 or 8 μm thick
	White plastic mulching (WPM)	Mulching with white plastic film 7 or 8 μm thick

Data processing

In this study, WUE (kg ha⁻¹ mm⁻¹) was defined as follows:(1) where Y is the potato yield (kg ha⁻¹), and ET is the evapotranspiration (mm) reported in the study.

The data were analyzed using the meta-analysis method described by Hedges et al. (Hedges and Olkin 1985). The effect size of plastic mulching for each observation of yield, WUE, and ET was estimated according to the SMD (standardized mean difference, g):(2) (3) (4) (5) where and are the means for the treatment and control groups, respectively; Swithin is the comprehensive within-group standard deviation for every study; NE and NC are the sample sizes for the treatment and control groups, respectively; SE and SC are the standard deviations for all comparisons in the treatment and control groups, respectively; and Vg is the variance of independent research.

The meta-analysis was performed using RevMan 5.1 software (developed by the Cochrane Collaboration, Oxford, UK), and in the present study, all of the variables of interest were continuous (Liu et al. 2011; Wang and Shangguan 2015; Xu, Liu, et al. 2015; Xu et al. 2016). Heterogeneity was assessed by both a chi-square test and an I² test (Wang and Shangguan 2015; Xu, Liu et al. 2015). The inverse-variance statistical approach was adopted for the meta-analysis. The differences in the means between the plastic-mulching and no-mulching groups were weighted based on their sample sizes and standard deviations, as determined by the RevMan program, and the confidence intervals (CI) were calculated according to their weighted effects. If the 95% CI values of the effect size for a variable did not overlap with zero, the effect size was considered significantly different. Statistical analyses were performed using SPSS statistical software (Version 20.0 for Windows, SPSS, Chicago, USA), and differences were evaluated at the 0.05 level of significance.

Results

Overview of the dataset

The potato yield and WUE were in the range of 9975–47,588 kg ha⁻¹ and 32.9–138.3 kg ha⁻¹ mm⁻¹, respectively, under plastic mulching. Seasonal ET was in the range of 142.8–442.6 mm under plastic mulching, but the mean values of 308.4–388.3 mm showed a relatively narrow range (Table 2).

Table 2. Actual values of potato yield, water use efficiency (WUE), and evapotranspiration (ET) under plastic mulching on the Loess Plateau.

Item	Categorical variable	Yield (kg ha^−1) range	WUE (kg ha^−1 mm^−1) range	ET (mm) range
Mulching method	FM	11300–43845	74.3–111.4	275.1–432.9
	CTFM	9600–30248	41.5–90.7	285.1–439.9
	RFM	9975–47588	32.9–138.3	142.8–442.6
	CTRFM	9600–34991	24.3–103.2	216.7–439.9
Mulching area	FAM	10575–47588	36.5–138.3	235.3–435.4
	CTFAM	9600–34991	24.3–103.2	253.8–439.9
	PAM	9975–45060	32.8–130	142.8–442.6
	CTPAM	9600–34991	24.3–103.2	216.7–439.9
Color of plastic film	BPM	14371–47588	36.5–138.3	202–442.6
	CTBPM	10583–34991	24.3–103.2	219–439.9
	WPM	9975–45060	32.9–116.4	142.8–436.1
	CTWPM	9600–33199	24.3–90.7	216.7–436.1
Precipitation	< 400 mm	9975–47422	59.3–138.3	142.8–413.7
	CT< 400 mm	9600–34991	31.2–103.2	216.7–385.6
	> 400 mm	13394–47588	32.9–122.2	235.3–442.6
	CT> 400 mm	10583–33199	24.3–90.7	306.5–439.9

Notes: FM: flat mulching, RFM: ridge-furrow mulching, FAM: full-area mulching, PAM: partial-area mulching, BPM: black plastic mulching, and WPM: white plastic mulching.

Among the 648 observations analyzed in the meta-analysis, 340 involved potato yield; 154 involved potato WUE; and 154 involved seasonal ET (Figure 3). In Figures 3(a) and (b), most of the points fall above the 1:1 line, indicating that the potato yield and WUE markedly increased under plastic mulching compared with those values for no mulching. For seasonal ET (Figure 3(c)), the points are distributed almost equally on both sides of the 1:1 line, demonstrating that the positive and negative effects were nearly equal and indicating that there was no significant difference in the effects of plastic mulching and conventional practices on seasonal ET.

Figure 3. Results of the comparison of plastic mulching and conventional tillage. (a) Potato yield, (b) water use efficiency (WUE), and (c) evapotranspiration (ET). In the Figure, each black point represents the result of one individual comparison between plastic mulching and the control. The points lying on the 1:1 line indicate that the corresponding values responded similarly to plastic mulching and the control, whereas the points distributed above or below the line indicate positive or negative effects, respectively, under plastic mulching.

Overall effects of plastic mulching on potato yield, WUE and ET

The overall effect sizes for potato yield and WUE were 1.57 (1.32–1.82) and 1.18 (0.79–1.58), respectively, and the 95% CIs for these parameters did not include zero, indicating a significant positive effect of plastic mulching on yield and WUE (Figure 4(a) and (b), Table 3). However, the overall effect size for ET (0.23, −0.05 to 0.51) indicated no notable difference between plastic mulching and conventional tillage (Figure 4(c), Table 3).

Figure 4. The relative effect sizes for potato yield, water-use efficiency (WUE), and evapotranspiration (ET) under different plastic mulching methods, mulching areas, colors of plastic mulch and annual precipitation on the Loess Plateau. Error bars represent the 95% CI, and values close to the bars represent the corresponding numbers of observations. FM: flat mulching, RFM: ridge-furrow mulching, FAM: full-area mulching, PAM: partial-area mulching, BPM: black plastic mulching, and WPM: white plastic mulching.

Table 3. Results of the meta-analysis (heterogeneity analysis) of the effects of plastic mulching on the yield, water-use efficiency (WUE), and evapotranspiration (ET) of potato on the Loess Plateau using random-effects models.

Item		Categorical variable	Number of observations	P-value of effect size	Heterogeneity analysis
					chi-square	df	P-value of chi-square test	Heterogeneity (I^2) (%)
Yield	Mulching method	FM	46	0.0002	26.52	13	0.23	17
		RFM	294	< 0.0001	175.75	40	0.05	17
	Mulching area	FAM	164	< 0.0001	103.81	32	0.04	22
		PAM	176	< 0.0001	103.14	39	0.11	16
	Color of plastic film	BPM	150	< 0.0001	82.65	21	0.23	10
		WPM	190	< 0.0001	115.91	31	0.06	19
	Precipitation	< 400 mm	104	< 0.0001	78.17	17	0.009	35
		> 400 mm	236	< 0.0001	121.99	26	0.36	4
WUE	Mulching method	FM	18	0.72	8.35	3	0.4	4
		RFM	136	< 0.0001	93.27	19	0.02	28
	Mulching area	FAM	82	< 0.0001	51.37	14	0.11	22
		PAM	72	< 0.0001	52.25	15	0.03	33
	Color of plastic film	BPM	92	< 0.0001	51.47	11	0.24	13
		WPM	62	0.0002	52.92	11	0.006	43
	Precipitation	< 400 mm	52	0.002	50.45	10	0.002	50
		> 400 mm	102	< 0.0001	53.51	9	0.34	7
ET	Mulching method	FM	18	0.62	6.52	3	0.59	0
		RFM	136	0.06	86.64	19	0.05	23
	Mulching area	FAM	82	0.74	45.72	14	0.25	13
		PAM	72	0.12	49.12	15	0.06	29
	Color of plastic film	BPM	92	0.8	34.45	11	0.87	0
		WPM	62	0.03	56.56	11	0.002	47
	Precipitation	< 400 mm	52	0.9	41.46	10	0.02	40
		> 400 mm	102	0.1	52.12	9	0.39	4

Notes: FM: flat mulching, RFM: ridge-furrow mulching, FAM: full-area mulching, PAM: partial-area mulching, BPM: black plastic mulching, and WPM: white plastic mulching.

On average, the potato yield under plastic mulching (27614 kg ha⁻¹) increased by 29.2% compared with that of conventional tillage (21,375 kg ha⁻¹) (Figure 5(a)). Similarly, the potato WUE under plastic mulching (84.5 kg ha⁻¹ mm⁻¹) increased by 31.7% compared with that of conventional tillage (64.2 kg ha⁻¹ mm⁻¹) (Figure 5(b)). However, there was almost no difference in the average ET under plastic mulching (361.3 mm) and conventional tillage (357.8 mm) (Figure 5(c)).

Figure 5. Potato yields (a), WUE (b), and ET (c) under plastic mulching and conventional tillage.

Effects of plastic mulching under different mulching methods

The effects of plastic mulching varied with the mulching method. The mean effect size of plastic mulching for the potato yield was 1.64 (1.37–1.91) under RFM and 1.18 (0.56–1.8) under FM (Figure 4(a)), so the RFM method caused a greater improvement in yield (30.0%) than FM (24.8%) (Figure 6(a)). The effect size for the potato WUE under RFM (1.33, 0.9–1.76) was significant, but the effect size for potato WUE under FM (0.17, −0.74 to 1.08), whose 95% CI included zero, was not significant (Figure 4(b)). The WUE increased under RFM by 32.5%, representing better performance than under FM, 26.4% (Figure 7(a)). The mean effect sizes for potato ET under RFM (0.3, −0.01 to 0.60) and FM (−0.2, −0.98 to 0.58) were not significantly different, and the average seasonal ET under RFM (364.1 mm) was higher than that under FM (340.1 mm) (Figure 8(a)).

Figure 6. Yields under plastic mulching and conventional tillage. (a) Mulching method, (b) mulching area, (c) color of plastic film, and (d) precipitation. FM: flat mulching, RFM: ridge-furrow mulching, FAM: full-area mulching, PAM: partial-area mulching, BPM: black plastic mulching, and WPM: white plastic mulching.

Figure 7. Water-use efficiency (WUE) under plastic mulching and conventional tillage. (a) mulching method, (b) mulching area, (c) color of plastic film, and (d): precipitation. FM: flat mulching, RFM: ridge-furrow mulching, FAM: full-area mulching, PAM: partial-area mulching, BPM: black plastic mulching, and WPM: white plastic mulching.

Figure 8. Evapotranspiration (ET) under plastic mulching and conventional tillage. (a) Mulching method, (b) mulching area, (c) color of plastic film, and (d) precipitation. FM: flat mulching, RFM: ridge-furrow mulching, FAM: full-area mulching, PAM: partial-area mulching, BPM: black plastic mulching, and WPM: white plastic mulching.

Effects of plastic mulching with different mulching extents

The dataset was separated into two sub-datasets according to different mulching extents. The mean effect size for the potato yield under FAM (1.73, 1.34–2.12) was larger than that under PAM (1.45, 1.12–1.77) (Figure 4(a)), and the increase in the yield under FAM (32.9%) was greater than that under PAM (25.7%) (Figure 6(b)). The mean effect size for potato WUE under FAM (1.28, 0.64–1.92) was slightly larger than that under PAM (1.11, 0.62–1.6) (Figure 4(b)), and potato WUE benefited more from FAM (35.7%) than from PAM (26.2%) (Figure 7(b)). Neither the effect size for potato ET under FAM (0.06, −0.29 to 0.41) nor that under PAM (0.38, −0.1 to 0.86) were significant (Figure 4(c)), and the average ETs under FAM (360.8 mm) and PAM (361.8 mm) were approximately the same (Figure 8(b)).

Effects of plastic mulching with different colors of plastic film

The effects of plastic mulching were also affected by the color of the plastic film. The mean effect size for the potato yield under BPM (1.34, 1.01–1.68) was smaller than that under WPM (1.74, 1.39–2.08) (Figure 4(a)). The mean effect size for potato WUE under BPM (0.95, 0.54–1.37) was also smaller than that under WPM (1.67, 0.78–2.55) (Figure 4(b)), and the increase in potato WUE under BPM (30.4%) was smaller than that under WPM (33.8%) (Figure 7(c)). The effect on potato ET under BPM (−0.04, −0.26 to 0.33) was not significant, but the corresponding value under WPM (0.72, 0.07–1.38) indicated a significant positive effect (Figure 4(c)).

Effects of plastic mulching under different levels of rainfall

The effect of rainfall < 400 mm (1.84, 1.21–2.47) on the potato yield was larger than the effect of rainfall > 400 mm (1.46, 1.21–1.71) (Figure 4(a)), and similarly, the increase in yield under rainfall < 400 mm (31.3%) was greater than the increase in yield under rainfall > 400 mm (28.4%) (Figure 6(d)). The effect on potato WUE under rainfall < 400 mm (2.68, 0.97–4.4) was much larger than the effect under rainfall > 400 mm (0.95, 0.6–1.3) (Figure 4(b)). There was greater improvement (47.4%) in potato WUE under rainfall < 400 mm than under rainfall > 400 mm (23.7%) (Figure 7(d)). Neither the effect on potato ET under rainfall < 400 mm (0.05, −0.67 to 0.77) nor the effect under rainfall > 400 mm (0.24, −0.04 to 0.53) were significant (Figure 4(c)). The average seasonal ET under rainfall >400 mm (388.3 mm) was much greater than that under rainfall <400 mm (308.4 mm) (Figure 8(d)).

Input, output and net revenue of potato

Among 44 publications included in this synthesis, 10 publications reported the economic benefits of potato under plastic mulching. The economic benefit was much higher in plastic film mulching than no mulching despite its higher cost in mulching material and labor (Table 4). The output/input and net revenue were increased by 10.86% and 34.43% for plastic mulching compared to no mulching.

Table 4. Analyses on economic benefits of potato.

Item	Output (RMB/ha)	Input (RMB/ha)	Net revenue (RMB/ha)	Output/Input (O/I)	Increase in Net revenue (%)	Increase in O/I (%)
No mulching	26296	7515	18781	3.50
Plastic mulching	34005	8758	25247	3.88	34.43	10.86

Water consumption characteristics

On the Loess Plateau, the relationship between the potato yield and seasonal ET under plastic film mulching could be described by a single quadratic function obtained via regression analysis after pooling all of the datasets (y = −0.9344x² + 624.0280x – 71469.2079, P < 0.01, Figure 9(a)). A curvilinear ET-yield relationship was also found under no mulching (y = −0.5374x² + 386.5037x – 43957.3058, P < 0.05, Figure 9(a)). The potato yield decreased when ET exceeded a certain critical value, which was approximately 334 mm under plastic film mulching and 360 mm under no mulching (Figure 9(a)). Regression analysis also produced parabolic WUE-ET relationships under plastic mulching (y = −0.0017x² + 1.0013x – 44.6867, P < 0.01) and no mulching (y = −0.0016x² + 0.9472x – 61.8693, P < 0.05) (Figure 9(b)). The relationship between WUE and yield is shown in Figure 9(c); potato WUE was observed to increase linearly with increasing yield under both plastic film mulching and no mulching. On the Loess Plateau, an increase in yield of 1 kg ha⁻¹ led to an increase in WUE of 0.0021 kg ha⁻¹ mm⁻¹ under plastic mulching and an increase of 0.0019 kg ha⁻¹ mm⁻¹ under no mulching (y = 0.0021x + 26.0003, R² = 0.7348, P < 0.01 and y = 0.0019x + 25.326, R² = 0.555, P < 0.01).

Figure 9. Relationships among the yield, water-use efficiency (WUE), and evapotranspiration (ET) of potatoes under plastic mulching on the Loess Plateau. (a) Relationship between ET and yield, (b) relationship between ET and WUE, and (c) relationship between yield and WUE.

Discussion

Responses of yield and water-use efficiency to plastic mulching

Our dataset consisted of 648 observations from 44 studies (Figure 1) conducted at 17 sites on the Loess Plateau (Table 1, Figure 2). Plastic mulching showed promoting effects for yield and WUE of potato compared with no mulching (Figure 3). In an analysis of wheat and maize yields under soil mulching, Qin et al. (Qin et al. 2015) found that plastic mulching increased both the yield and WUE of wheat by 20% and increased both the yield and WUE of maize by 60%. The meta-analysis conducted in the present study demonstrated that plastic mulching had significant positive effects on the potato yield and WUE on the Loess Plateau (Tables 2 and 3, Figure 4); our results based on data collected from the peer-reviewed literature showed that plastic mulching increased potato yield and WUE by 29.2% and 31.7%, respectively (Figure 5). Plastic film mulching can decrease soil bulk density and increase both soil porosity and available nutrient contents (Hou et al. 2015), decrease soil evaporation, enhance soil water infiltration and redistribute soil moisture, thereby relieving water stress to some degree (Li, Wang, et al. 2004; Gan et al. 2013). In addition to modifying the soil micro-environment, plastic film mulching can increase the photosynthetic capacity and the endogenous hormone levels of potato plants (Hou et al. 2015). Due to these many advantageous effects, plastic film mulching increases potato yields both quantitatively and qualitatively and enhances WUE (Qin et al. 2014; Zhao et al. 2014). Our study also showed that there was no significant difference in ET under plastic mulching and no mulching (Figures 4, 5 and 8). The results are consistent with previous study of wheat on the Loess Plateau (Wang and Shangguan 2015).

Wang and Shangguan (Wang and Shangguan 2015) studied the effects of different mulching and tillage practices on wheat yields on the Loess Plateau and found that, among the five mulching practices examined, RFM was the most effective practice for improving wheat yields. In the present study, similar results were obtained for potatoes, another crop that is widely planted in the same region (Figure 4). The increases in yield and WUE under RFM were higher than those under FM (Figures 6 and 7). RFM can enhance the topsoil temperature early in the growing season (spring) when temperatures are low, and this higher soil temperature is essential for seed germination (Gan et al. 2013; Hu et al. 2014). Tian et al. (Tian et al. 2003) reported that the average runoff efficiencies (expressed as the ratio of runoff to rainfall) under three mulched ridge treatments (mulched ridges with widths of 0.3, 0.45 and 0.6 m) were 53.1%, 63.3% and 69.7%, respectively, when the average precipitation was 223 mm. Zhao et al. (Zhao et al. 2012) showed that RFM was associated with a high rainfall harvesting efficiency of more than 85% when the average precipitation was 382 mm. Due to this efficiency, especially in a region with little rainfall, RFM retains more stored water and increases the amount of water available to plants than FM (Tian et al. 2003), and the higher temperatures and humidity in the topsoil create favorable conditions for seed germination and seedling growth. Increased emergence rates and strong seedling establishment result in vigorous growth and high potato yields. In addition, RFM can increase the receiving surface area for solar radiation, raise the temperature on a ridge during the day and increase the cooling area at night, thereby increasing the difference in soil temperature between day and night, which is conducive to crop growth.

Our results suggest that FAM leads to better performance than PAM on the Loess Plateau (Figure 4). Full mulching with plastic film can greatly improve tuber yields and WUE in comparison with no mulching because the increase in yield and WUE decreases as the mulching area decreases (Figures 6 and 7). For example, in Dingxi (Gansu, China) in 2009, FAM increased the potato yield and WUE by 57.4% and 70.3%, respectively, compared to the yield and WUE of the control group, while half mulching increased the potato yield and WUE by only 33.1% and 31.9%, respectively; in early growing stages, FAM raised the soil temperature to a much greater extent than half mulching (Hu et al. 2014). Full mulching has also been shown to decrease soil water evaporation, increase the penetration of rainwater in the soil, and contribute to the accumulation of soil moisture (Ramakrishna et al. 2006). Additionally, the combined benefits of higher topsoil moisture and temperatures under FAM not only improve the potato emergence rate but accelerate the speed of germination (Zhao et al. 2012; Hu et al. 2014). Plant height, the leaf area index, the relative growth rate and the dry matter content have also been shown to be higher under FAM than under PAM (Hu et al. 2014; Qin et al. 2014), and FAM was found to prolong the growth period from seedling to maturation, thus contributing to better tuber development. Therefore, FAM leads to higher potato yields and WUE than PAM.

Studies on colored plastic film mulch have provided different or even contradictory results. Chang et al. (Chang et al. 2016) reported that potato grown from the clear plastic mulch produced more total and >80 g yields than with black plastic mulch. Ibarra et al. (Ibarra-Jimenez et al. 2011) found that the use of black plastic mulch resulted in a lower total yield than lighter mulch colors (aluminum, white or silver) in Mexico. Zhang et al. (Zhang, Wang, et al. 2017) reported that potato grown with white plastic mulch had higher WUE than potato grown with black plastic mulch under drip irrigation in Gansu Province of Northwest China. Similar result was also found in cucumber cultivation (Yaghi et al. 2013). In contrast, Singh and Ahmed (Singh and Ahmed 2008) reported that the tuber yield under black polythene mulching was higher than under white polythene mulching; emergence, plant height, and the number of stems were also improved with black polythene mulching. Anikwe et al. (Anikwe et al. 2007) found that black plastic mulched plots provide superior edaphic environment and higher yield for cocoyam when compared to other treatments used. For the 44 publications included in this synthesis, mulching practices in 22 publications were white plastic mulch, 12 publications were black plastic mulch and 10 publications (one publication reported that potato yield of white plastic mulch was significantly higher than that of black plastic mulch, three publications showed the contrary results and other six found that there were no significant difference between the two colors) contained both colors. Because we could not obtain a comprehensive result directly from the original literature, therefore we used meta-analysis and calculated the effect sizes to conduct quantitative comparison of the two colors. WPM was found to perform better than BPM on the Loess Plateau (Figure 4), which may be because the mean annual temperature at most sites in this region is lower than 10°C, which constrains crop growth. Additionally, the light transmissivity of white plastic film is greater than that of black plastic film; thus, most of the solar radiation can pass through white plastic film, be directly absorbed by the soil and raise its temperature, thereby affecting crop growth. It is possible that the effect of WPM on increasing temperature is better than that of BPM, thus WPM results in a higher potato yield than BPM.

Highly effective collection and utilization of rainfall is crucial to increasing agricultural productivity in semi-arid regions, where the crop yield generally depends on rainfall. Qin et al. (Qin et al. 2015) synthesized data regarding water input levels and found that the mean effect of plastic mulching on maize yield was 60% with low water input (< 370 mm) and 40% with high water input (> 370 mm), while the mean effect of plastic on the maize WUE was 70% and 40%, respectively. Our results showed that the increases in the potato yield and WUE associated with plastic mulching under rainfall < 400 mm were greater than those under rainfall > 400 mm compared with those with no mulching (Figures 6 and 7), which is consistent with the results of Zhou (Zhou et al. 2009) showing that plastic mulching exhibited great potential to increase crop production under limited rainfall. Similarly, plastic mulching significantly increased the maize yield by 15–26% in dry years, but no significant increase yield was observed in rainy years in northeast China (Xu, Li, et al. 2015). Such yield increases in dry years or dry regions resulted from a greater rate of dry matter accumulation due to the higher topsoil temperature and water content under plastic mulching.

Economic benefit of plastic mulching

In this study, the output value of plastic film mulching based on tuber yield was considerably higher than that of no mulching. Compared to no mulching, plastic mulching required additional plastic film and labor. However, no mulching also required more labor for weeding. The input value of plastic mulching was higher than that of no mulching. The resulting increase in output value offset the increase in preharvest costs and assured a positive net revenue. Consequently, the net revenue was increased by 34.43% for plastic mulching in comparison to no mulching (Table 4). Similar results were also found in other crops such as maize, cabbage and cotton (Dong et al. 2009; Paranhos et al. 2016; Zhang, Wei, et al. 2017).

Relationships among potato yield, WUE, and ET under plastic mulching

Wang and Shangguan (Wang and Shangguan 2015) summarized data for dryland wheat from 1995 to 2014 and found a quadratic relationship between wheat yields and seasonal ET. Our study indicated a quadratic yield-ET relationship for potato under plastic mulching and no mulching (Figure 9(a)), and similar relationships have been found for winter wheat under limited irrigation and long-term nitrogen fertilization regimes in northwest China (Kang et al. 2002; Zhong and Shangguan 2014). The potato yield-ET relationship under plastic mulching described in this study has not been previously reported, and the equations shown in Figure 9(a) demonstrated that potatoes required an ET of 334 and 360 mm to achieve the highest yield under plastic mulching and no mulching, respectively. Previous studies have indicated that curvilinear relationships exist between wheat WUE and seasonal ET on the Loess Plateau (Kang et al. 2002; Zhong and Shangguan 2014; Wang and Shangguan 2015), but there have been no relevant studies of potato. In the present study, significant parabolic relationships between potato WUE and ET were observed under plastic mulching and no mulching (Figure 9(b)). When ET was relatively low, water availability was the limiting factor for yield, and an increase in ET resulted in a significant increase in both yield and WUE. However, the augmentation of yield and WUE began to decrease as ET further increased. For the yield-ET relationship, the curve for plastic mulching fell above that for no mulching, demonstrating that with the same ET, a higher yield could be obtained under plastic mulching than under no mulching. Plastic mulching was associated with a higher attainable yield than no mulching, and a similar WUE-ET relationship existed under plastic mulching and no mulching. The analysis of the yield-ET and WUE-ET relationships provides a possible means of exploring the potential potato yield and the yield gap between conventional tillage and plastic mulching on the Loess Plateau. Consistent with findings for dryland wheat in the same region, potato WUE increased linearly as yield increased under both plastic mulching and no mulching (Figure 9(c)). In addition, the slopes (ratios of WUE and yield) for potatoes under plastic mulching and no mulching were 0.0021 and 0.0019, respectively, which are similar to the corresponding value (0.002) for wheat (Wang and Shangguan 2015). The yield-WUE line for plastic mulching fell above that for no mulching, which indicated that a higher WUE could be achieved under plastic mulching than under no mulching when the yield was equal. Our results demonstrated that a higher WUE could generally be obtained with a higher yield. A higher WUE means that the crop can obtain a higher yield with less water, so increasing WUE is important for balancing the conflict between a higher yield demand and a lower water supply in rain-fed areas.

Limitations of the meta-analysis and future expectations

The data presented in this synthesis were derived from a screening of the literature regarding field experiments on the Loess Plateau involving different plastic mulching methods, mulching extents, colors of plastic film and levels of precipitation, and heterogeneities may have derived from the following factors: (1) the researchers had different personal experiences and preferences; (2) many of the field experiments did not include any long-term observations; (3) the experimental fields exhibited different nutrient statuses and fertilizer application rates before sowing; and (4) climate conditions and field management practices differed during the potato-growing season. Although the effect of plastic mulching can be evaluated via the meta-analytical method, the underlying sources of meta-analytical uncertainties require further research.

Some effective planting patterns associated with plastic film mulching have been implemented in recent years to increase potato productivity and WUE on the Loess Plateau (Zhao et al. 2012; Hu et al. 2014; Qin et al. 2014), but the development of more appropriate field management strategies remains difficult in this region. Using a combination of different field management practices, such as tillage, mulching, fertilization, and intercropping and crop rotation, crop yields and WUE may be further improved in the future. Such integrated measures will not only decrease fluctuations in crop productivity but will provide diverse environmental benefits.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes on contributors

Qiang Li is from the Institute of Soil and Water Conservation, Yangling, China. His major research directions include yield potential of dryland potato.

Hongbing Li is from the Institute of Soil and Water Conservation, Yangling, China. The major research directions include ecological plant physiology.

Suiqi Zhang is from the Institute of Soil and Water Conservation, Yangling, China. The major research directions include crop eco-physiological adaptation to arid environment and field crops management.

Additional information

Funding

This work was supported by National Science and Technology Supporting Programs [grant number 2015BAD22B01]; 111 project of the Chinese Education Ministry [grant number B12007]; and the Special Funds for Scientific Research Programs of the State Key Laboratory of Soil Erosion and Dryland Farming on the Loess Plateau [grant number A314021403-C5].