Soil fertility, leaf nutrients and their relationship in kiwifruit orchards of China’s central Shaanxi province

ABSTRACT

Understanding soil fertility and leaf nutrients provides the basis for optimizing fertilization for fruit crop production. However, for central Shaanxi province of China, the largest kiwifruit-producing region in the world, limited information is available on soil fertility and leaf nutrients of these orchards. Here, we assessed 17 soil fertility parameters and 12 leaf elements for 116 kiwifruit orchards in central Shaanxi. The investigated region included five locations: Zhouzhi, Mei, Yangling, Wugong and Qishan. The soils in this region had an average pH of 7.54 and exchangeable Ca of 5.68 g/kg, indicating that the soils were calcareous. Over a quarter of the orchards were deficient in soil organic matter. Although over 50% of the soils were excessive in NO₃^–-N and over 25% both deficient and excessive in P and K, more than 65.0% of the orchards were deficient in leaf N, P and K, implying the low NPK use efficiency in central Shaanxi. Moreover, Cl and Zn deficiencies as well as Cu and Ca excesses were found in both soil and leaf diagnosis. Analysis of variance showed that, among the study locations, Zhouzhi was characterized by low soil pH, low soil/leaf Ca and high soil/leaf N, whereas Wugong was featured by high soil/leaf Na. Correlation analysis indicated that soil pH positively correlated with soil Ca, and both parameters negatively correlated with soil N, P, Fe, Mn, Cu and Cl (P < 0.01). Additionally, soil N, Ca, Mn and Cl positively correlated with the corresponding leaf elements (P < 0.001). These results could be used to guide the sustainable nutrient management of kiwifruit production in central Shaanxi.

KEY WORDS:

• Kiwifruit
• soil fertility
• leaf nutrient
• calcareous soil
• nutrient diagnosis

1. Introduction

Kiwifruit (Actinidia Lindl.) is one of the most important recently domesticated fruit crops in the world. In 2015, the kiwifruit yield and harvested area in China represented 2,187,867 tons and 181,900 ha respectively, and they accounted for 52.7% and 69.0% of the global kiwifruit yield and area (FAO 2018). In China, most kiwifruit is produced in Shaanxi, Sichuan, Henan, Anhui, Jiangxi, Hunan, and Hubei provinces (Huang 2014), and the crop in central Shaanxi represents 47.1% and 24.3% of China’s yield and area, respectively (Zhang 2016; FAO 2018). In recent years, the amount of kiwifruit planted in central Shaanxi has increased due to the possible high profits from kiwifruit production and the supportive measures provided by local agricultural policy. However, these kiwifruit orchards are primarily run by local smallholders, with an average area of 0.19 ha, and they are often characterized by poor management practices (Lai 2011). These orchardists often fail to apply fertilizers rationally because of a lack of scientific fertilization knowledge and awareness of environmental protection, and because of their pursuit of economic benefits. An investigation of the Yujiahe catchment in Zhouzhi county of central Shaanxi province showed that 36.4% of kiwifruit orchards received no manure applications and 94.3% had insufficient manure input; moreover, 81.8% of orchards suffered from excess N fertilizer application, and excess and deficient application of P and K fertilizers was also a problem (Lu et al. 2016a). Incorrect fertilization not only has the potential to result in reduced plant growth, fruit quantity and quality, but it can also result in increased production costs and potential environmental risk as well as food safety issues (Müller et al. 2015; Gao et al. 2016; Lu et al. 2016b; Zhao et al. 2017).

Understanding soil fertility and leaf nutrients is the basis for optimizing fertilization, and it is very important for improving kiwifruit quantity and quality (Lai 2011; Lago et al. 2015; Zhao et al. 2017). To date, there has been considerable research on kiwifruit nutrient management (e.g., Smith et al. 1987a; Battelli and Renzi 1990; Coutinho and Veloso 1997; Zhang et al. 2001; Tarakcioglu et al. 2007; Mahmoudi et al. 2014; Khachi et al. 2015). In China, the soil fertility and leaf nutrients in kiwifruit orchards have been investigated in Shaanxi (Zhang et al. 2001; Li et al. 2008; Yu et al. 2011; Lai 2011; Kang 2014; Lu et al. 2016a), Sichuan (Pan et al. 2014), Jiangxi (Huang et al. 2014), Hubei (Xu et al. 2011), and Guizhou (Wan et al. 2015). However, there is still room for improvement because (1) most soil fertility research is limited to assessing the soil pH, organic matter, and NPK, while little attention has been paid to the micronutrient (Fe, Mn, Cu, Zn, B, and Cl) and salt contents, which are nevertheless crucial for maintaining the balance of nutrients and for the sustainable production of perennial crops such as kiwifruit; (2) studies on leaf nutrients have primarily focused on ‘Qinmei’ and ‘Hayward’ but barely on any of the other currently predominant cultivars, such as ‘Xuxiang’ and ‘Cuixiang’; (3) only a few studies on the relationship between soil fertility and leaf nutrients in kiwifruit orchards have been conducted (Xu et al. 2011; Huang et al. 2014); and (4) the sample size is often too small to reflect the local soil fertility and leaf nutrient well. Therefore, it is still necessary to investigate the soil fertility and leaf nutrients in kiwifruit orchards in central Shaanxi province on a large scale and in relation to multiple important parameters.

The objectives of this study were (1) to assess the soil fertility and leaf nutrient status of the kiwifruit orchards in central Shaanxi, through analyzing 17 soil parameters (pH, organic matter, salt, N, NO₃^–-N, NH₄⁺-N, P, K, Ca, Mg, Fe, Mn, Cu, Zn, B, Cl and Na) and 12 leaf elements (N, P, K, Ca, Mg, Fe, Mn, Cu, Zn, B, Cl and Na) of 116 orchards including 5 cultivars (‘Hayward’, ‘Qinmei’, ‘Xuxiang’, ‘Cuixiang’ and ‘Huayou’), and (2) to analyze the relationship between soil properties and leaf nutrients in this region.

2. Materials and methods

2.1. Site description

The study sites were located in the central region of Shaanxi province (34°06ʹ～34°27ʹN, 106°12ʹ～108°18ʹE, evaluation 323～800 masl), China. This area has sufficient sunshine and a warm and temperate continental monsoon climate with hot, wet summers and cold, dry winters. The average annual temperature and precipitation ranges are 11 to 13°C and 500 to 800 mm, respectively. Approximately 60% of the precipitation occurs between June and September. The three primary landscapes are represented by the orchards surveyed in this study as follows: an alluvial fan at the foot of the Qinling Mountains, a high terrace, and Wei River Plain. The soils are calcareous and are classified as calcic kastanozems (FAO 2015) or earth-cumuli-orthic anthrosols or cultivated loessial soils according to the China Soil Taxonomy (Xiong and Li 1987).

2.2. Sampling

In late August and early September of 2016, extensive soil and leaf sampling was performed at the following two regions: (1) north of the Qinling Mountains and south of the Wei River (including 39 samples from Zhouzhi county, and 58 samples from Mei county), and (2) north of the Wei River (including 6 samples from Yangling, 7 samples from Wugong county, and 6 samples from Qishan county). In total, 116 samples (i.e., one sample per orchard) were collected. The investigated cultivars included ‘Xuxiang’ (Actinidia chinensis var. deliciosa ‘Xuxiang’; 59 orchards), ‘Hayward’ (A. chinensis var. deliciosa ‘Hayward’; 35 orchards), ‘Qinmei’ (A. chinensis var. deliciosa ‘Qinmei’; 9 orchards), ‘Huayou’ (A. chinensis var. chinensis ‘Huayou’; 7 orchards), and ‘Cuixiang’ (A. chinensis var. deliciosa ‘Cuixiang’; 6 orchards). Most of the samples were collected from Zhouzhi and Mei because over 65% of the kiwifruit production in Shaanxi province takes place in these two counties (Zhang 2016).

Soil samples were taken at a depth of 0 to 30 cm and a distance of 1.0 to 1.2 m from the main vines with a soil auger of 5 cm interior diameter (Huang et al. 2014). Each soil sample was composed of five random cores per orchard, stored in double-lined plastic bags, and brought to the laboratory for soil property analysis. The soils were air-dried, ground by a rolling pin to pass through a 1.00-mm sieve (half of the soil sub-samples were passed through a 0.15-mm sieve for organic matter analysis) and stored in sealed plastic bags until analysis.

Leaf samples were taken from the second leaf past the final fruit cluster on a fruiting lateral, and each sample consisted of 40 leaves (blade plus petiole) from 20 vines with 2 leaves per vine (Smith et al. 1987a). The samples were washed in sequence with tap water, 0.5% detergent solution, 0.1% hydrochloric acid (HCl) solution, and deionized water. The leaf samples were then oven-dried at 65°C to a constant weight, ground in a mill to pass through a 20-mesh sieve and stored in sealed plastic bags for analysis.

When the leaf and soil samples were taken, data on the yield and cultivation-related information for each orchard were obtained from the growers and reassessed by a local kiwifruit-selling agent who is an expert at judging kiwifruit yield to guarantee the reliability of the orchard information.

2.3. Laboratory analysis

Fourteen parameters of the soil chemical properties (i.e., pH, organic matter (OM), salt, P, K, Ca, Mg, Fe, Mn, Cu, Zn, B, Cl, and Na) were analyzed according to the methods described by Bao (2000), and N content was determined according to the method described by Lu et al. (2016b). In brief, soil pH was measured with a glass electrode in a 1:5 suspension of soil:solution, and electrical conductivity (EC) was measured in the same extract and the total salt concentration was calculated by EC. Organic matter content was determined by the potassium dichromate volumetric method using external heating. Mineral N (including ammonium N (NH₄⁺-N) and nitrate N (NO₃^–-N)) was extracted with 1.0 M KCl and determined using a continuous flow analyzer (Seal AA3, Norderstedt, Germany). Available soil P was determined by molybdate blue method after extraction with NaHCO₃. Exchangeable cations (K⁺, Ca²⁺, Mg²⁺, and Na⁺) were extracted with 1.0 M NH₄OAc, and most available micronutrients (Fe²⁺, Mn²⁺, Cu²⁺, and Zn²⁺) were extracted with DTPA extractant containing 0.05 M diethylenetriaminepentaacetic acid (DTPA), 0.01 M CaCl₂ and 0.1 M triethanolamine (TEA), and all were determined by an atomic absorption spectrophotometer (Hitachi Z-2000, Tokyo, Japan; Bao 2000). Available B was extracted with boiling water and determined by an inductively coupled plasma optical emission spectrometry (ICP-OES; Thermo IRIS-Advan, Massachusetts, USA; Bao 2000). Water-soluble Cl was extracted from a 1:2.5 soil-water mixture and determined by a discontinuous analyzer (DeChem-Tech.GmbH CleverChem200, Hamburg, Germany; Bao 2000).

The concentrations of P, K, Ca, Mg, Fe, Mn, Cu, Zn, B and Na in plant leaves were determined by ICP-OES, and the Cl was determined with the discontinuous analyzer after the samples were dry-ashed (with the addition of CaO only for Cl measurement) and dissolved in 0.5 M HNO₃ solution (Zhou et al. 2014). The N concentration in leaf tissue was quantified by Kjeldahl method (Bao 2000).

2.4. Statistical analysis

Unless otherwise stated, all the data were computed using Microsoft Office Excel2007. For each given parameter, the deficiency frequency (%) represented the percentage of the sample number below optimum range in the total sample number (n = 116); the optimum frequency (%) represented the percentage of the sample number in optimum range in the total sample number (n = 116); and the excess frequency (%) represented the percentage of the sample number above optimum range in the total sample number (n = 116). The data in Tables 2, 4 represented the means of individual orchards (replicates) in each location ± standard deviations (SDs), and differences among locations were separated by Duncan’s multiple range test at P < 0.05 using SPSS for Windows 16.0 (SPSS Inc., Chicago, IL, USA). Pearson’s correlation in Fig. 3 was tested at P < 0.05 and the Fig. 3 was made using the corrplot R library in R version 3.2.2 (R Foundation for Statistical Computing, Vienna, Austria).

Table 1. Soil fertility status of kiwifruit orchards in central Shaanxi province (n = 116)†

	Mean	CV (%)	Sample range	Optimum range	Deficiency frequency (%)	Optimum frequency (%)	Excess frequency (%)
pH	7.54	9.1	4.87–8.28	5.5–6.5	1.7	6.9	91.4
OM (%)	1.66	15.8	1.04–2.60	1.5–4.0	25.9	74.1	0.0
Salt (g/kg)	0.80	44.3	0.27–2.54	<1.0	-	78.4	21.6
N (mg/kg)	54.24	69.7	12.87–220.50	-	-	-	-
NO3^––N (mg/kg)	42.97	82.6	3.95–184.22	11.0–28.0	12.1	35.3	52.6
NH4^+-N (mg/kg)	11.26	54.6	1.70–43.58	-	-	-	-
P (mg/kg)	57.55	73.8	5.04–195.73	30.0–60.0	26.7	41.4	31.9
K (mg/kg)	279.52	47.5	98.81–715.52	250.0–300.0	50.9	15.5	33.6
Ca (g/kg)	5.68	26.4	1.03–7.51	1.2–3.6	0.9	11.2	87.9
Mg (mg/kg)	245.28	24.9	99.85–419.46	120.0–360.0	1.7	94.8	3.5
Fe (mg/kg)	7.24	245.3	0.78–112.93	4.5–10.0	81.9	5.2	12.9
Mn (mg/kg)	5.64	101.8	0.93–36.57	7.0–12.0	76.7	13.8	9.5
Cu (mg/kg)	1.06	52.4	0.37–3.55	0.5–1.0	4.3	55.2	40.5
Zn (mg/kg)	0.86	109.8	0.08–5.64	1.0–2.0	73.3	19.0	7.8
B (mg/kg)	0.73	28.5	0.38–1.45	0.5–1.0	12.1	79.3	8.6
Cl (mg/kg)	15.24	104.9	3.42–98.40	10.0–30.0	50.9	39.7	9.5
Na (mg/kg)	26.15	69.3	5.29–118.10	<115.0	-	99.1	0.9

† The optimum range is adopted from Li et al. (1996; NO₃^––N), Bao (2000; Salt), Zhang et al. (2004; Na), Tong (2011; Fe, Mn, Cu and B), Huang et al. (2014; pH, OM, P, Ca, Mg and Cl), Zhang et al. (2016; K) and Yu (2017; Zn).

Table 2. The comparison of five locations in soil fertility parameters of kiwifruit orchards from central Shaanxi province†

	Zhouzhi	Mei	Yangling	Wugong	Qishan
pH	7.01 ± 0.73b	7.86 ± 0.30a	7.87 ± 0.05a	8.05 ± 0.14a	6.85 ± 1.13b
OM (%)	1.70 ± 0.29a	1.67 ± 0.24a	1.71 ± 0.12a	1.41 ± 0.15b	1.53 ± 0.27ab
Salt (g/kg)	0.95 ± 0.46a	0.71 ± 0.27a	0.76 ± 0.08a	0.97 ± 0.22a	0.64 ± 0.06a
N (mg/kg)	81.24 ± 42.81a	42.34 ± 28.24b	27.68 ± 7.39b	47.23 ± 24.31b	28.42 ± 8.76b
NO3^–-N(mg/kg)	67.26 ± 39.26a	32.04 ± 28.11b	20.70 ± 8.89b	38.58 ± 25.52b	18.22 ± 4.73b
NH4^+-N (mg/kg)	13.98 ± 8.07a	10.30 ± 4.41ab	6.98 ± 2.96ab	8.65 ± 2.56b	10.19 ± 5.79ab
P (mg/kg)	86.83 ± 48.48a	46.29 ± 31.99b	42.25 ± 23.60b	30.42 ± 16.81b	23.09 ± 11.58b
K (mg/kg)	329.24 ± 146.83a	259.36 ± 127.53ab	287.59 ± 69.43ab	243.29 ± 107.11ab	185.42 ± 42.88b
Ca (g/kg)	4.66 ± 1.69c	6.16 ± 0.99ab	6.74 ± 0.25a	6.70 ± 0.22a	5.34 ± 2.22bc
Mg (mg/kg)	257.43 ± 40.24bc	225.70 ± 57.23cd	292.91 ± 41.91b	340.83 ± 70.35a	196.41 ± 65.82d
Fe (mg/kg)	13.77 ± 23.96ab	2.78 ± 3.68b	2.07 ± 0.71b	1.28 ± 0.47b	20.02 ± 43.58a
Mn (mg/kg)	8.95 ± 7.83a	3.88 ± 3.26ab	5.20 ± 2.09ab	1.80 ± 0.51b	6.06 ± 4.74ab
Cu (mg/kg)	1.46 ± 0.70a	0.85 ± 0.30bc	0.99 ± 0.18bc	0.58 ± 0.14c	1.05 ± 0.42ab
Zn (mg/kg)	0.96 ± 0.97ab	0.88 ± 0.96ab	1.24 ± 1.37a	0.28 ± 0.13b	0.32 ± 0.11b
B (mg/kg)	0.81 ± 0.14bc	0.63 ± 0.16d	1.17 ± 0.24a	0.84 ± 0.09b	0.68 ± 0.29cd
Cl (mg/kg)	23.68 ± 21.61a	11.38 ± 11.00ab	6.71 ± 2.00b	14.19 ± 4.87ab	7.40 ± 4.90b
Na (mg/kg)	24.80 ± 8.01c	19.78 ± 6.76c	36.72 ± 13.62b	84.74 ± 25.58a	17.48 ± 7.80c

† Values are means ± SDs. Different letters indicate significant differences between five locations for each given parameter at P < 0.05. Zhouzhi, n = 39; Mei, n = 58; Yangling, n = 6; Wugong, n = 7; and Qishan, n = 6.

Table 3. Leaf nutrient status of kiwifruit orchards in central Shaanxi province (n = 116)†

	Mean	CV (%)	Sample range	Optimum range	Deficiency frequency (%)	Optimum frequency (%)	Excess frequency (%)
N (g/kg)	21.4	10.9	14.8–27.3	22.7–27.7	69.0	31.0	0.0
P (g/kg)	1.4	21.1	0.8–2.5	1.6–2.0	73.3	20.7	6.0
K (g/kg)	9.6	38.0	3.1–23.2	12.0–18.6	76.7	21.6	1.7
Ca (g/kg)	40.8	13.7	26.7–56.4	32.9–44.3	6.0	69.0	25.0
Mg (g/kg)	7.6	32.4	0.6–12.0	4.0–11.3	6.0	90.5	3.4
Fe (mg/kg)	174.4	33.4	94.8–457.4	90.1–267.9	0.0	94.8	5.2
Mn (mg/kg)	138.4	76.1	36.7–604.6	44.5–173.1	0.9	81.9	17.2
Cu (mg/kg)	34.9	24.9	18.9–66.2	7.0–21.8	0.0	4.3	95.7
Zn (mg/kg)	26.1	44.1	11.3–89.9	23.6–44.2	51.7	40.5	7.8
B (mg/kg)	55.1	14.8	40.2–80.7	38.5–79.9	0.0	99.1	0.9
Cl (g/kg)	4.2	30.6	2.4–11.4	6.0–10.0	94.0	5.2	0.9
Na (mg/kg)	178.4	18.9	113.7–277.4	<500.0	-	100.0	0.0

† The optimum range is adopted from Zhang et al. (2001; N, P, Ca, Mg, Fe, Mn, Cu, Zn, B and Cl), Kang (2014; K) and Smith et al. (1987a; Na).

Table 4. The comparison of five locations in leaf nutrients of kiwifruit orchards from central Shaanxi province†

	Zhouzhi	Mei	Yangling	Wugong	Qishan
N (g/kg)	23.21 ± 1.76a	20.86 ± 2.02b	19.96 ± 2.25bc	18.83 ± 2.44c	19.98 ± 1.03bc
P (g/kg)	1.43 ± 0.25a	1.46 ± 0.33a	1.39 ± 0.29a	1.55 ± 0.42a	1.33 ± 0.23a
K (g/kg)	10.27 ± 3.71ab	9.23 ± 3.63ab	10.62 ± 3.00a	6.88 ± 0.90b	11.14 ± 4.70a
Ca (g/kg)	37.39 ± 4.67c	42.79 ± 5.00ab	39.76 ± 3.66bc	41.15 ± 4.90abc	45.22 ± 8.01a
Mg (g/kg)	6.04 ± 2.49c	8.57 ± 1.99ab	7.42 ± 1.84bc	10.17 ± 1.35a	6.24 ± 1.25c
Fe (mg/kg)	150.86 ± 36.60c	170.09 ± 61.09bc	215.56 ± 18.63ab	250.85 ± 32.32a	238.26 ± 61.83a
Mn (mg/kg)	171.89 ± 147.68a	117.54 ± 62.33a	97.21 ± 17.02a	110.30 ± 35.57a	195.54 ± 149.39a
Cu (mg/kg)	34.80 ± 7.62ab	34.49 ± 8.54ab	40.90 ± 12.46a	30.91 ± 4.73b	38.44 ± 14.35ab
Zn (mg/kg)	22.43 ± 7.51b	26.72 ± 13.31b	39.05 ± 9.09a	25.27 ± 7.05b	32.34 ± 9.93ab
B (mg/kg)	54.94 ± 9.04a	55.57 ± 8.01a	58.82 ± 5.42a	52.03 ± 5.11a	51.27 ± 7.47a
Cl (g/kg)	4.80 ± 1.66a	3.86 ± 0.97ab	3.43 ± 0.46b	4.24 ± 0.60ab	3.98 ± 0.71ab
Na (mg/kg)	178.88 ± 26.89ab	180.54 ± 38.09ab	163.78 ± 17.70bc	199.43 ± 18.45a	145.59 ± 36.10c

† Values are means ± SDs. Different letters indicate significant differences between five locations for each given element at P < 0.05. Zhouzhi, n = 39; Mei, n = 58; Yangling, n = 6; Wugong, n = 7; and Qishan, n = 6.

3. Results

3.1. Evaluation of soil fertility in the kiwifruit orchards of Central Shaanxi

We collected 116 soil samples from the kiwifruit orchards and analyzed 17 parameters representing the soil fertility status. Most of the soils were characterized by high pH and Ca, with average values of 7.54 and 5.68 g/kg respectively (Table 1), indicating that the soils in central Shaanxi’s kiwifruit orchards were calcareous. Moreover, 52.6%, 40.5% and 21.6% of the orchards had excess soil NO₃^–-N, Cu and salt, respectively (Table 1). In contrast, 25.9% of the orchards were low in OM, and over 50.0% of the soil samples were deficient in Fe, Mn, Zn and Cl (Table 1). Additionally, 31.9% and 33.6% of the soils were excessive but 26.7% and 50.9% of the soils were deficient in P and K respectively (Table 1).

Soil pH in Zhouzhi and Qishan was lower than those in Mei, Yangling and Wugong (Table 2). Likewise, soil Ca in Zhouzhi was lower than those in Mei, Yangling and Wugong, with that in Qishan being medium (Table 2). In contrast, soil N, NO₃^–-N and P were higher in Zhouzhi than in the other four locations (Table 2). Additionally, the Wugong’s orchards contained the greatest contents in soil Na and Mg (Table 2). The distribution pattern of each soil parameter at deficiency, optimum and excess levels for the five locations coincided with the statistical results mentioned above (Fig. 1 and Table 2).

Figure 1. Soil fertility status of kiwifruit orchards in five locations from central Shaanxi province. Zhouzhi, n = 39; Mei, n = 58; Yangling, n = 6; Wugong, n = 7; and Qishan, n = 6

3.2. Evaluation of leaf nutrients in the kiwifruit orchards of Central Shaanxi

To assess the leaf nutrient status of these orchards, we further analyzed 12 elements of 116 leaf samples from the same orchards. We found that 69.0%, 73.3% and 76.7% of the orchards were deficient in leaf N, P and K, respectively (Table 3). Moreover, 94.0% and 51.7% of the leaf samples were deficient in Cl and Zn (Table 3). By contrast, 95.7% and 25.0% of the orchards showed excessive leaf Cu and Ca (Table 3).

Zhouzhi had the highest leaf N concentration, followed by Mei, Qishan and Yangling, and then Wugong (Table 4). Moreover, the orchards had higher leaf Cl concentration in Zhouzhi than in Yangling, with the other three locations being medium (Table 4). In contrast, leaf Ca concentration of Zhouzhi was lower than that of Mei and Qishan, and that of Yangling and Wugong was intermediate (Table 4). In addition, Wugong had a higher concentration of leaf Na than Yangling and Qishan, with Zhouzhi and Mei being medium (Table 4). The distribution pattern of each leaf element at deficiency, optimum and excess levels for the five locations agreed well with the statistical results of leaf nutrient concentrations (Fig. 2 and Table 4).

Figure 2. Leaf nutrient status of kiwifruit orchards in five locations from central Shaanxi province. Zhouzhi, n = 39; Mei, n = 58; Yangling, n = 6; Wugong, n = 7; and Qishan, n = 6

3.3. Relationship between fruit yield, soil fertility and leaf nutrients for kiwifruit orchards in Central Shaanxi

To clarify the relationship among these investigated parameters, we analyzed the correlations of fruit yield, soil fertility, leaf nutrients, and soil-leaf parameters for 116 pairs of the kiwifruit orchards, and we found that the amount and percentage of significant correlation were the highest for soil fertility variables (87 out of 136, i.e., 64.0%), followed by soil-leaf parameters (72 out of 204, i.e., 35.3%), leaf nutrients (18 out of 66, i.e., 27.3%) and yield-soil/leaf parameters (3 out of 29, i.e., 10.3%; Fig. 3). Among them, the kiwifruit yield positively correlated with soil P and leaf B, but negatively correlated with soil Na in central Shaanxi’s orchards (P < 0.05; Fig. 3).

Figure 3. Pearson correlations between the fruit yield, soil fertility and leaf elements of kiwifruit orchards in central Shaanxi province. ‘N.l’ indicates the leaf N and the same for other elements. The hollow dark rectangles indicate significant correlations between the same element for soil and leaf, including N, Ca, Mn and Cl. The red and blue color of the circles indicate the positive and negative sign of the correlations, respectively. Both the size and the color shade of the circles are proportional to the correlation coefficient value. Only significant correlations (P < 0.05, r = 0.182) are displayed and the non-significant correlations are left blank. n = 116

Except for soil pH and Ca, most of the soil fertility parameters showed positive correlations with one another (Fig. 3). Soil N had a higher correlation with NO₃^–-N (r = 0.99, P < 0.001) than with NH₄⁺-N (r = 0.45, P < 0.001; Fig. 3). Moreover, soil pH had an extremely significant positive correlation with soil Ca (r = 0.80, P < 0.001), and both pH and Ca negatively correlated with several nutrients (Fe, Mn, Cu, Cl, N and P; P < 0.01; Fig. 3). Similarly, a very high correlation was also found between soil salt and soil N (r = 0.89, P < 0.001), and both parameters strongly correlated with soil Cl (r = 0.62 and 0.66 respectively, P < 0.001; Fig. 3).

Leaf N, Ca, Mn and Cl were positively related to the corresponding elements in the soils (P < 0.001; Fig. 3). Moreover, leaf N negatively correlated with soil pH and Ca, but it positively correlated with soil salt, N (including NO₃^–-N and NH₄⁺-N), P, Fe, Mn, Cu and Cl (P < 0.01; Fig. 3). A similar trend in leaf N was also observed in leaf Cl and Mn (P < 0.05; Fig. 3). However, leaf Ca positively correlated with soil pH and Ca, but it negatively correlated with soil N (NO₃^–-N), Fe, Mn, Cu and B (P < 0.05; Fig. 3). A similar trend in leaf Ca was also found in leaf Mg (P < 0.05; Fig. 3).

Leaf N had a negative correlation with leaf Ca, Mg and Fe, but it had a positive correlation with leaf K, Mn and Cl (P < 0.05; Fig. 3). Moreover, leaf K was negatively related to leaf Mg, but positively related to leaf N, P, Fe, Mn, Zn and Cl (P < 0.05; Fig. 3). In addition, leaf Ca positively correlated with leaf Mg (P < 0.05; Fig. 3).

4. Discussion

Central Shaanxi is one of the most important kiwifruit-growing areas, accounting for 24.8% of the world’s yield (Zhang 2016; FAO 2018), and our investigation could provide important information for guiding kiwifruit fertilization and production in this region, thereby contributing to the sustainability of the kiwifruit industry.

4.1. N, P and K of kiwifruit orchards in Central Shaanxi

As macronutrients, N, P and K play vital roles in the nutritional diagnosis of soil fertility and leaf elements (Smith et al. 1987a). The disorders of N, P and K in kiwifruit cultivation have been reported in several countries, such as New Zealand (Smith et al. 1987a), Portugal (Coutinho and Veloso 1997), China (Zhang et al. 2001), Turkey (Tarakcioglu et al. 2007) and Iran (Mahmoudi et al. 2014). In central Shaanxi, excess of soil N and both deficiency and excess of soil P and K were identified (Table 1). These results were supported by a field experiment carried out in Yangling in which applying high P and K compound fertilizer with 60% as base fertilizer in last autumn and 40% as topdressing in current year improved the kiwifruit yield and quality more significantly than applying high N compound fertilizer with the same method (Yang 2016). However, the higher soil N (NO₃^–-N) and P values were observed in Zhouzhi when compared with the other four locations (Table 2), in agreement with previous studies performed in Zhouzhi (Tian et al. 2004; Lu et al. 2016a). Therefore, reducing N, balancing P and K according to specific locations is necessary.

Although both deficiency and excess of soil P and K as well as excess of soil N existed, over 65.0% of the orchards were deficient in leaf N, P and K (Tables 1, 3). The discrepancies in diagnostic results between soil and leaf nutrients may be mainly attributed to low use efficiency of NPK (particularly of N) in central Shaanxi’s kiwifruit orchards, likely as a consequence of nutrient leaching (Gao et al. 2016), low soil OM and high soil pH (Table 1) as well as arid climate in this region (Xiong and Li 1987). Therefore, increasing NPK use efficiency will be a challenge in kiwifruit cultivation of central Shaanxi in the future. Another possible explanation for these discrepancies is the errors induced by different study cultivars and sampling time in leaf diagnosis (Zhang et al. 2001; Kang 2014). For example, the sampling time of the present study was one month later than that of Zhang et al. (2001) that the optimum range was used, which might lead to lower N, P and K concentrations in leaves of our study than those of Zhang et al. (2001). Therefore, the optimum ranges of leaf nutrient for new kiwifruit cultivars and various growth stages in central Shaanxi should be developed to guide optimized fertilization. Noteworthy, the optimum range of soil mineral N has not been established in this region. Considering that the high correlation between soil mineral N and NO₃^–N (r = 0.99, P < 0.001; Fig. 3) and the probability distribution of NO₃^––N concentrations at deficiency, optimum and excess levels, here we propose 21.4–40.0 mg/kg as optimum soil mineral N range for central Shaanxi’s kiwifruit orchards. However, further field experiments are still needed to test its validity.

4.2. Cl, Zn, Cu, Mn, and Fe of kiwifruit orchards in Central Shaanxi

Micronutrient disorders have been observed in several countries, such as New Zealand (Asher et al. 1984), Greece (Sotiropoulos et al. 1998), Italy (Rombolà et al. 2000) and China (Liu et al. 2002). Our results showed that the kiwifruit orchards in central Shaanxi were deficient in Cl and Zn but they displayed excessive Cu in both soil and leaf evaluation (Tables 1, 3). These results are consistent with previous studies (Zhang et al. 2001; Yang 2016). These nutrient deficiencies may be attributed to (1) a large amount of nutrient removal through harvested fruit over many years (Ferguson and Eiseman 1983; Vajari et al. 2018), (2) soil nutrient depletion after the long-term selective absorption of ions from the same location by fruit trees, and (3) nutrient leaching to a soil depth that most roots are unable to reach (Lu et al. 2016b; Yang 2016). For example, kiwifruit species have a high requirement for Cl (Smith et al. 1987b), and Cl deficiency is frequently observed during kiwifruit production (Zhang et al. 2001; Tarakcioglu et al. 2007; Parent et al. 2015). A recent field experiment in central Shaanxi showed that applying 170 to 227 kg/ha of Cl-containing fertilizer increased the kiwifruit yield by 26.5 to 14.4% and raised the fruit soluble solid content when compared with the control treatment (Yang 2016). These results imply that applying Cl-containing fertilizers during kiwifruit production may be viable and necessary in this area. Additionally, 95.7% of the orchards were excessive in leaf Cu while only 40.5% of the orchards were excessive in soil Cu (Tables 1,3). Similarly, 81.9% of the samples were optimum in leaf Mn whereas 76.7% of the orchards were deficient in soil Mn (Tables 1,3). These results may be associated with the use of the Cu/Mn-containing agricultural chemicals in above-ground plant parts during kiwifruit cultivation. Besides, the excess of soil Cu may be explained by the high application dose of the Cu-containing agricultural chemicals in the canopy that the chemical residues often drip on the soil surface.

Kiwifruit is very sensitive to Fe deficiency (Tagliavini and Rombolà 2001), and leaf yellowing induced by Fe deficiency appears frequently in central Shaanxi (Liu et al. 2002; Yao et al. 2005; Tran et al. 2012). However, our results showed that the orchards in central Shaanxi were adequate in terms of their total leaf Fe despite their soil Fe deficiency (Tables 1,3). This result may be attributed to the fact that the total leaf Fe is not a reliable indicator for kiwifruit Fe diagnosis (Mahmoudi et al. 2014), but the active leaf Fe is more effective for Fe diagnosis than the total Fe (Mao et al. 2002; Wang et al. 2011; Liu et al. 2017). Nevertheless, no optimum range for active Fe has been developed for kiwifruit nutrient diagnosis to date. Here, we have suggested three approaches to mitigate Fe deficiency during kiwifruit production. These approaches could be either (1) the use of tolerant rootstock or scions, (2) compensating for Fe through the combined application of FeSO₄ or synthetic vivianite with manure (Rombolà et al. 2003), or (3) intercropping with graminaceous plants (Xiong et al. 2013).

4.3. Relationship between soil fertility and leaf nutrients in calcareous soils

Soil is a mineral reserves storing various nutrients for the normal growth of higher plants, and thus soil fertility parameters correlate closely with one another (Rahman et al. 2011; Huang et al. 2014). Our results showed that the greater amount and percentage of significant correlation were found between soil fertility variables (87 out of 136, i.e., 64.0%) than between soil-leaf parameters (72 out of 204, i.e., 35.3%), leaf nutrients (18 out of 66, i.e., 27.3%) and yield-soil/leaf parameters (3 out of 29, i.e., 10.3%; Fig. 3). Among the soil-leaf parameters, leaf N, Ca, Mn and Cl were found to be positively related to the corresponding soil nutrients (P < 0.001; Fig. 3), inconsistent with only Cl (Xu et al. 2011) and B (Huang et al. 2014) soil-leaf correlations reported in acidic soils, probably as a consequence of the differences in soil types.

Our investigation indicated that soil Ca was positively related to soil pH (r = 0.80, P < 0.001; Fig. 3), implying that soil Ca might act in regulating soil pH (Xiong and Li 1987). Furthermore, soil pH negatively correlated with several soil elements (Fe, Mn, Cu, Cl, N and P; P < 0.01; Fig. 3) and leaf nutrients (Mn, N, K and Cl; P < 0.01; Fig. 3). However, soil pH was positively related to leaf Ca and Mg (P < 0.001; Fig. 3). These results suggest that maintaining an appropriate and stable soil pH is central to the sustainable soil nutrient management of kiwifruit orchards in central Shaanxi. Our results showed that the soil pH in this region was as high as 7.54, and 91.4% of the orchards showed high soil pH (Table 1). The application of chemical fertilizers seems to be able to decrease the soil pH (Guo et al. 2010; Lu et al. 2016b). However, overfertilization will lead to a series of problems, such as the deterioration of the soil physio-chemical properties (e.g., weakening the soil buffer capacity), the decline of microbial biomass (Carey et al. 2009), soil salinization and nutrient leaching (e.g., NO₃⁻, Cl⁻, and Ca²⁺; Gao et al. 2016; Yang 2016; Li et al. 2017), thereby resulting in low nutrient use efficiency, the spread of disease and pests, potential environmental risk and food safety issues (Guo et al. 2010; Lu et al. 2016b). For example, the soil N (NO₃^–-N) and P contents in Zhouzhi were higher than those in the other four locations (Table 2). However, long-term overfertilization in Zhouzhi orchards has led to soil N excess and leaching as well as slight soil salinization (Tian et al. 2004; Gao et al. 2016; Lu et al. 2016a; Fig. 1). In another example, the soils of Mei county’s kiwifruit orchards seemed to be acidifying because in 2006, the soil pH was 8.19 ± 0.24 (Li et al. 2008), but in 2015 and 2016, the values were 7.83 ± 0.36 (Fan 2017) and 7.86 ± 0.30 (Table 2). Alternatively, applying manure and biofertilizer to increase the soil OM content may be a promising way to enhance the capacity to orchestrate soil pH (Li et al. 2012; Ku et al. 2018). It has been reported that organic fertilizers play a crucial role in improving the kiwifruit yield, quality and growth performance (Khachi et al. 2015; Lago et al. 2015; Zhao et al. 2017). However, in our study, 25.4% of the soils in central Shaanxi were deficient in OM (Table 1). Hence, enhancing the application of organic fertilizers is highly necessary for kiwifruit cultivation in this region (Lu et al. 2016a). In addition, organic management for kiwifruit orchards that precludes the use of synthetic fertilizers and herbicides but permits the application of mineral-bearing rocks, plant residues, and animal manures (Carey et al. 2009; Lago et al. 2015) might be introduced in central Shaanxi in the future.

Based on our results and those of previous studies, here we proposed an overview of the relationship between soil-leaf parameters for kiwifruit vines grown on calcareous soils that predominate in central Shaanxi (Xiong and Li 1987; Fig. 4). This overview might help us to better understand the characteristics of soil fertility and leaf nutrients in central Shaanxi, thereby contributing to the sustainability of the kiwifruit industry.

Figure 4. Overview of the relationship between soil fertility and leaf nutrients for kiwifruit orchards in central Shaanxi province. The soil and leaf parameters are represented as filled yellow and green rectangles, respectively. Green lines indicate positive effects and red lines indicate negative effects; the thicker the lines are, the stronger the effects are

5. Conclusion

Our results showed that the kiwifruit orchards in central Shaanxi were characterized by high soil pH and Ca, indicating that the soils were calcareous. The investigated orchards exhibited low organic matter, disorders of soil N, P and K, and deficiencies of leaf N, P and K, as well as deficiencies of Zn and Cl in both soil and leaf. Therefore, on the whole, replenishing soil organic matter and micronutrients (e.g., Zn and Cl), controlling soil P and K, and reducing soil N are necessary for the sustainable management of kiwifruit orchards in central Shaanxi. For Zhouzhi’s orchards, controlling chemical fertilizer application to avoid soil salinization/acidification and enhancing organic fertilizer application to improve soil physio-chemical properties are highly necessary; while for Wugong’s orchards, remedying the limiting nutrients and elevating the OM content are promising strategies. However, further fertilizer-response experiments based on the current study are still needed to guide the recommendation of fertilizer application rates and types.

Acknowledgments

We are very grateful to Ms Linping Hu and Ms Mengyuan Li from the College of Horticulture, Northwest A&F University, for participating in the soil analysis. This work was funded by the National Natural Science Foundation of China (31601710), the Scientific Startup Foundation for Doctors of Northwest A&F University (Z109021611), and the Fundamental Research Funds for the Central Universities (Z109021603).

Additional information

Funding

This work was supported by the National Natural Science Foundation of China [31601710]; Fundamental Research Funds for the Central Universities [Z109021603]; Scientific Startup Foundation for Doctors of Northwest A&F University [Z109021611]