Effects of Seven Diversified Crop Rotations on Selected Soil Health Indicators and Wheat Productivity

Diversified cropping systems can enhance soil condition and increase system productivity worldwide. To reduce the negative effects that accompany the continuous winter wheat–summer maize (WM) double-cropping in the North China Plain (NCP), diversified crop rotation (DCR) needs to be considered. The objective of this study is to evaluate the effect of DCR on soil health and wheat productivity as compared to a continuous WM double-cropping. A field experiment (37°41′ N, 116°37′ E) was established in the NCP including a traditional WM double-cropping as a baseline. During 2016/2017–2017/2018, the control is winter wheat–summer maize→winter wheat–summer maize (WM→WM) and seven DCRs as follow: fallow→winter wheat–summer maize (F→WM); spring maize→winter wheat–summer maize (Ms→WM); winter wheat→winter wheat–summer maize (W→WM); sweet potato→winter wheat–summer maize (Psw→WM); spring peanut→winter wheat–summer maize (Pns→WM); winter wheat–summer peanut→winter wheat–summer maize (WPn→WM) and potato–silage maize→winter wheat–summer maize (PMl→WM). Our results indicated that DCRs significantly changed certain soil health indicators in 2016/2017 compared with the control, where F→WM rotation significantly decreased soil pH by 2.7%. The DCRs, especial Psw→WM and Pns→WM rotations showed a potential positive effect on soil health indicators at the end of the second year (2017/2018) compared with the control, where sweet potato increased soil organic carbon (SOC), total nitrogen (TN), available phosphorus (AP), urease activity (UA) and alkaline phosphatase activity (APA) in 2017/2018 by 5.1%, 5.3%, 13.8%, 9.4%, and 13.5%, respectively. With the spring peanut, TN, AP, and soil APA were increased by 2.1%, 13.2%, and 7.7%, respectively. Although fertilizer and irrigation input of DCRs were lower than the control, no significant decrease was observed on actual wheat yield as compared to the control (7.79 Mg/ha). The finding of this study highlights the value of DCRs, especially, Psw→WM and Pns→WM rotations over WM double-cropping in the NCP.


Introduction
Meeting the growing food demand of the increasing population with limited agricultural resources is a major challenge on national and global scales [1,2]. The world population is forecasted to reach about 9.6 billion by 2050, while the world's cultivated land area has increased by only 12% [3,4]. Meanwhile, crop rotations can significantly affect selected soil health indicators at the end of the rotation-year compared with traditional rotation, and (2) wheat productivity among the DCRs will maintain stability or decrease slightly with lower fertilizer and irrigation input compared with the control.

Experiment Site
A new eight crop rotations experiment established in October 2015 at Wu Qiao Experiment Station (37 • 41 N, 116 • 37 E) of China Agricultural University, Hebei province, Northern China. The station is in warm, semi-humid and continental temperature monsoon zone, with an annual average temperature of 12.9 • C and an annual average rainfall of 562 mm, which concentrated during June to August and a 201 day frost-free period. The soil at the experiment site is classified as Calcaric Fluvisol [35] with a sandy clay loam texture. The baseline soil samples were collected in September 2016 for 0-20 cm depths, which include: soil bulk density (1.40 Mg/m 3 ), soil pH (8.13), soil organic carbon stock (13.49 Mg/ha), and soil total nitrogen stock (2.29 Mg/ha). Prior to the establishment of the study, the typical continuous double-cropping is winter wheat→summer maize (WM) with one-time conventional chisel plow tillage per year before winter wheat sowing.

Experiment Design
During 2016/2017-2017/2018 ( Figure 1, Table 1), the control is winter wheat-summer maize→winter wheat-summer maize (WM→WM) and seven DCRs as follow: fallow→winter wheat-summer maize (F→WM), spring maize→winter wheat-summer maize (Ms→WM), winter wheat→winter wheat-summer maize (W→WM), sweet potato→winter wheat-summer maize (Psw→WM), spring peanut→winter wheat-summer maize (Pns→WM), winter wheat-summer peanut→winter wheat-summer maize (WPn→WM), potato-silage maize→winter wheat-summer maize (PMl→WM). The experiment includes wheat and maize planted in 2017/2018 (Table 1) and other crops mentioned above planted in 2016/2017 (Table 1) in three replications (n = 3) in 30 m 2 experiment plot size in a complete randomized block design. The growing crop and corresponding season of each year and different rotation, fertilizer and irrigation requirements for each crop in this study were summarized from October 2016 to October 2018 (Table 1). During the experiment, we used chisel plow tillage before the first crop sowing in that year. After crop harvest, crop residue is incorporated into the topsoil of each plot using chisel plow tillage except for silage maize, which is removed for livestock use.
Agronomy 2020, 10, x FOR PEER REVIEW 3 of 14 will maintain stability or decrease slightly with lower fertilizer and irrigation input compared with the control.

Experiment Site
A new eight crop rotations experiment established in October 2015 at Wu Qiao Experiment Station (37°41′ N, 116°37′ E) of China Agricultural University, Hebei province, Northern China. The station is in warm, semi-humid and continental temperature monsoon zone, with an annual average temperature of 12.9 °C and an annual average rainfall of 562 mm, which concentrated during June to August and a 201 day frost-free period. The soil at the experiment site is classified as Calcaric Fluvisol [35] with a sandy clay loam texture. The baseline soil samples were collected in September 2016 for 0-20 cm depths, which include: soil bulk density (1.40 Mg/m 3 ), soil pH (8.13), soil organic carbon stock (6.74 Mg/ha), and soil total nitrogen stock (1.14 Mg/ha). Prior to the establishment of the study, the typical continuous double-cropping is winter wheat→summer maize (WM) with one-time conventional chisel plow tillage per year before winter wheat sowing.

Soil Sampling and Measurements
Soil samples were collected from all crop rotations prior to harvest in September 2017. Additional soil samples were collected from all rotations in mid-June 2018 prior to the winter wheat harvest. Soil samples were collected using a metal ring for the top 0-20 cm soil depth. Soil gravimetric water content (GWC) was determined by weighing the wet and dry soil samples, which were oven-dried at 105 • C overnight [36]. Soil samples for bulk density (BD) were taken with a metal ring and weighed wet, then dried and weighed again to determine soil bulk density [37,38]. Soil pH was determined with an FE20-K pH meter (FiveEasy, Shanghai, China) by mixing 10 g of soil with 25 mL of water (China agricultural standard, NY/T 1377-2007). Soil organic carbon (SOC) concentration was measured by Schellenberger method and soil samples were treated with H 2 SO 4 to remove inorganic C from soil [39]. Soil total N (TN) concentration was measured by the semi-micro Kjeldahl method [40] using automatic nitrogen analyzer (Foss Kjeldahl sampler 8400, Shanghai China). Soil available P (AP) concentration was measured by the Olsen method [41]. Soil organic carbon concentration was multiplied by depth (m) and soil bulk density (Mg/m 3 ) to be presented as soil organic carbon stocks (Mg/ha), and soil nitrogen concentration was multiplied by the same way to get soil nitrogen stocks (Mg/ha) [42]. Soil enzymatic activities such as sucrose and urease determined as soil biological indicators for soil health. Sucrase and urease activities determined by a modified method described by [42], and Ge et al. [43] to measure sucrase and urease activities in their research. Additionally, alkaline phosphatase activity was determined by using a method described by [44]. The same measurement was done by Liu et al. [45].
During five different wheat growth periods, which are: emerged, jointing, flowering, milky stage, and turning stage (152, 179, 194, 215, and 235 days after winter wheat emerged) of the growing season, samples of winter wheat were collected using a frame of 20 cm long by 15 cm wide of one row and taken to the laboratory for measurements. Then all the plant samples were dried at 65°C, then weighed for the total above-ground biomass dry weight [46]. Grain dry weight, spikes per plant, kernels per spike, and the weight of kernel were recorded at harvest time. Plants were hand-harvested by cutting them at ground level. Spikes and kernels were counted by hand, and grain was threshed with a single thresher. The kernels were dried and weighed with an electronic scale [47].

Data Analysis
The statistical analysis was done using the IBM SPSS.25. One-way analysis of variance (ANOVA) was employed to test the significance of mean differences (at p value of 0.05). A significant effect determined at p < 0.05 and means were separated using Duncan multiple-range procedure. The means of selected soil health indicators in 2016/2017 and 2017/2018 were analyzed using statistical analysis. The results summarized in Tables 2 and 3. Furthermore, in order to explore the effect of diversified crop rotations on soil health indicators, we calculated the rate of change of these indicators as compared to the control (WM→WM) in two years ( Figure 2).      Table 2).

Soil Physical Properties
According to the analysis of the BD and GWC rate of change compared with that of the control in 2017/2018 (Figure 2a), the growth trend is limited. For example, Ms→WM, Pns→WM, WPn→WM, and PMl→WM rotations decreased soil bulk density by 1.2%, 1.7%, 3.1%, and 2.1%, respectively. The GWC in F→WM, W→WM, and Psw→WM rotations all increased by less than 2% compared to that in the control in 2017/2018 (8.1%) ( Table 2 and Figure 2b).

Soil Chemical Properties
Diversified crop rotations had significant effects in 2016/2017, compared with the control on soil chemical indicators (Table 2). F→WM rotation (soil pH = 8.25) caused soil pH to decrease by 2.7% ( Table 2). Among all seven DCRs, Psw→WM rotation was the only one had significantly lower soil SOC, TN, and AP in 2016/2017 compared with the control, with a rate of decline of 17.6%, 18.7%, and 43.7%, respectively (Table 2) (Figure 2f).

Soil Biological Properties
In this study, most of the soil enzymes activities showed no significant change among all the rotations in two years with two exceptions in 2017/2018 ( Table 2). The F→WM and Psw→WM rotations caused a significant decrease in SA in 2016/2017 by 19.3% and 28.4%, respectively, compared to the control ( Table 2). In F→WM rotation, UA was 4.91 mg NH 3 -N g soil −1 d −1 in 20116/2017, which was significantly lower than that in the control, 6.26 mg NH 3 -N g soil −1 d −1 (Table 2).
Except for WPn→WM rotation, all other DCRs increased certain selected soil chemical indicators, soil pH, SOC, TN, and AP in 2017/2018, and W→WM, Psw→WM, and Pns→WM rotations had two out of three increases among SA, UA and APA. Although no significant change occurred in 2017/2018, Ms→WM and W→WM rotations increased SA by 6.5% and 4.8% over control (Figure 2g). UA decreased in most DCRs compared with the control (Figure 2i). In addition, W→WM, Psw→WM, Pns→WM, and PMl→WM rotations had greater APA than that in the control in 2018, with rate of increase: 4.3%, 13.5%, 7.7% and 2.6%, respectively.

Wheat Productivity
Diversified crop rotations showed no significant difference in the aboveground biomass production of winter wheat during different growth stages as compared to the control (Figure 3), while some significant differences observed in wheat yield parameters ( Table 3). The control has the significantly lowest kernel weight, 38.74 g per 1000-kernel compared to the other seven rotations, which range from 39.48-43.39 g per 1000-kernel. In other words, it means enhancing crop species diversity has beneficial potential for wheat productivity (Table 3). There were no significant differences in the final winter wheat yield among the other rotations compared with the control (Table 3).

Soil Health Indicators
Some significant changes in soil health indicators caused by crop rotations in 2016/2017 were observed at the end of that year. For instance, F→WM rotation significantly decreased soil pH by 2.7% compared to the control in the NCP, where lower soil pH is beneficial to crops in the NCP. The use of urea (CO(NH 2 ) 2 ) in this study shows no significant effect on soil pH [48,49]. In another research, it was found that soil pH of fallow-maize is lower than that of soybean-maize and cowpea-maize rotations [49]. Based on the analysis above, the use of the fallow system was beneficial in lowering soil pH in this study. In addition, Psw→WM rotation decreased SOC, TN, AP, and SA in this study in 2016/2017 as compared with the control in the short-term. In general, the soil SOC, TN, and AP under Psw→WM rotation were significantly lower than that in the control in 2017/2018. This can be attributed to the potential removal of considerable amount of nutrients from soil by sweet potato in the Psw→WM rotation [50]. However, the majority of these nutrients were returned into the soil after the sweet potato residues (leaves) were incorporated in the soil with plow tillage. This result concurs with the findings reported that returning crop straw increased soil total nitrogen and available phosphorus significantly at soil depth of 0-20 cm [20,51]. Therefore, it is reasonable to find that Psw→WM rotation had the significantly lower soil nutrients content and soil enzymes than the control.
Diversifying crop rotations had an effect on certain soil health indicators at the end of 2017/2018 compared with the traditional control, which is similar to other research results. More than half of the selected soil health indicators (physical, chemical or biological) were improved by the end of 2017/2018 under F→WM, W→WM, Psw→WM, and Pns→WM rotations in this study. For instance, F→WM, W→WM, and Psw→WM rotations showed higher SOC than that with the control in 2017/2018. Other studies documented that adding crop species improved the SOC [20,21]. Nonetheless, the SOC with PMl→WM rotation was always lower than that of the control. This may be attributed to maize biomass removal for animal feed instead of it returning it back to the field [31]. The DCRs showed a potential to decrease urease activity compared with the control. The TN with F→WM, W→WM, Psw→WM, Pns→WM, and WPn→WM rotations were higher than that in the control, WM rotation. Other studies documented similar findings with DCR [19].

Wheat Yield and Diversified Crop Rotation
Although the control had the lowest kernel weight than other DCRs, no significant wheat yield differences were observed among the seven DCRs as compared with the control. Based on this study's findings, our hypotheses regarding crop rotations effects in stabilizing wheat productivity with lower fertilizer and irrigation input of the seven DCRs as compared with control is supported by the findings of this study. However, other studies reported that diversifying crop rotations were beneficial in increasing the grain yield system productivity [15,16,18]. The potential reason is enhancing crop diversity has a potential benefit to improve wheat productivity that may occur in the long-term experiment. The compensation of cereals, i.e., to take advantage of favorable conditions throughout the crop life cycle, may form a particular balance among the yield components [52], which can lead to the significant different kernel weight, while being insignificant in the actual wheat yield in 2017/2018.
In this study, it has been demonstrated that there is a potential positive impact by DCR on selected soil health indicators and wheat productivity maintenance. These findings are similar to other studies, which show that DCRs are not only beneficial to soil health, but also to crop yield [33,[52][53][54]. However, further definitive evidence was lacking due to the experiment being in the early stage of the long-term period. Long-term observations under DCR are needed to document in greater detail the long-term impact of such rotations on soil health indicators and crop productivity.

Conclusions
The findings of this study demonstrated the potential positive impact of DCR on selected soil health indicators and wheat productivity, by using different crop species included in different rotations in the NCP. Diversified crop rotations significantly affected certain soil health indicators at the end of the first year (2016/2017) compared with the control, especially, F→WM and Psw→WM rotations. The majority of the seven DCRs, especially Psw→WM and Pns→WM rotations, showed potential effect at the end of the second year (2017/2018) of the study as compared with the control. The DCRs showed an upward trajectory for improving soil health indicators, such as, GWC, BD, SOC, SA, UA and APA during the early years of establishment of this study. These indicators are essential to improve the main crop productivity such as wheat in the North China Plain, which is seriously short of water and has decreased soil fertilizer condition.

Conflicts of Interest:
The authors declare no conflict of interest.