Effects of Straw Returning on Soil Physicochemical Properties and Microbial Diversity Under the Rice-craysh Integrated System

This study represents the investigation of soil physicochemical properties and microbial diversity by the Biolog ECO analysis in a 7-year eld experiment using winter ooded fallow + no straw returning (W), winter ooded fallow + straw returning (WS), and winter ooded fallow + straw returning + craysh farming (WSC) at soil depths of 0-10 cm and 10-20 cm. Compared with the W treatment, the WS treatment has signicantly more total reducing substances in the 0-10 cm layer and sucrase activity in the 0-20 cm layer. The WSC treatment has signicantly more available N (AN), total N (TN) contents, acid phosphatase and sucrase activities in the 10-20 cm layer, while the pH value, total reducing substances, and Fe 2+ content in the 0-20 cm layer was considerably lower compared with the WS treatment. Biolog ECO analysis reveals that microbial community composition in the WS and WSC treatments differs from that in the W treatment in the 0-20cm layer. The WS treatment increases signicantly the diversity of bacterial community and the ability of utilizing carbon sources in the 0-10 cm layer, and the species abundance of bacterial community in the 0-20 cm layer. Meanwhile, in the 10-20 cm soil layer, the WSC treatment improves the species abundance of bacterial community and the ability of utilizing carbon sources compared with the WS treatment. These results indicate that straw returning under the rice-craysh integrated system improves soil physicochemical properties, decreases reducing substances properties, and increases soil enzyme activity and functional diversity of microbial community, thereby contributing to soil condition. the soil reducing substance contents here. Meanwhile, in our study, the straw returning decreased signicantly the total reducing substances in the 0-10 cm layer. The incorporation of a mass of straw to the upper soils may be the reason, which can further deepen the anaerobic environment of the soil. Soil microbial community functional diversity is an indicator of soil microbe community structure and function, and reects the ecological characteristics of soil microbes 40 . The Biolog ECO method is a sensitive method for detecting functional changes in soil microbial community structure, and is broadly used to assess soil microbial community functional diversity 41 . In this study, we observed that the AWCD value of the WS and the WSC treatments always higher than those in the W treatment during all cultured time. Meanwhile, the WS treatment increased the soil microbial diversity, soil microbial richness, and the ability of utilizing carbon sources in the 0-20 cm layers. Those were in good agreement with those of Yu et al 42 . The reason for this was the straw returning contributed to the greater accumulation of rice residues on the soil surface, promoting the growth and proliferation of certain groups of microbes. Compared to the WS treatment, the WSC treatment increased signicantly the microbial utilization rate of carbon source and the species abundance of bacterial community in the 10-20 cm layer. Those were consistent with the results reported by Zhu et al 43 . It may due to the reason that the craysh activity could enhance the dissolved oxygen in water to reach the deep layer, change the availability of nutrients and oxygen, and affect microbial community structure, function, and diversity. Research on the utilization of the different carbon sources by soil microbe could reveal detailed information about microbial community metabolism 44 . In this study, the results showed that the straw returning and craysh farming had a greatly inuence on the community composition of soil microbe in the 0-20 cm layers. The straw returning can increased the microbial populations and the craysh farming alleviated the anaerobic situation of soil and increased aerobic bacterial community, which made the utilization of carbon sources for microbe increase at this time. Soil microbial diversity was affected by soil type, moisture, pH, and soil management measures 45 . The correlation analysis showed that the soil pH value, AK, TN, and Fe 2+ contents and total reducing substances were signicantly


Introduction
The double-cropping of cray sh (Procambarus clarkii) and rice (Oryza sativa) had been practiced in Louisiana, USA, which was introduced to China in the early 21th century, and then the rice-cray sh integrated system was formed [1][2][3] . In the rice-cray sh integrated system, the cray sh can be allowed to live in the paddy elds, where the rice straw served as the food basis for cray sh after the rice was harvested. This system fully utilized the shallow water environment and the winter idle period of rice paddies, combined planting and aquaculture industries, and really increased farmers' income. Thus, it was seen that the rice-cray sh integrated system was great economic and social bene ts. Currently, the rice-cray sh integrated system had become one of the primary cultivation models in middle and lower reaches of Yangtze River, and was performed in approximately 4.6 ×10 5 ha in the Hubei Province in 2019, China. It was signi cantly changed in the rice-cray sh integrated system to rice output, soil physicochemical properties, and the structure of soil microbial community 1,4-5 . Si et al 6 found that the rice-cray sh integrated system increased soil carbon levels, and strongly affected microbial community composition and structure in the deeper layers of soil. Yuan et al 7 showed that cray sh aquaculture in rice elds signi cantly improved the soil quality using a minimum data set.
It was reported that the total output of straw in China ranks rst in the world (~1.04 billion tons), with large straw returning area and wide regional distribution 8- 10 .Straw returning, a major way of utilizing straw resources, can directly or after stacking and decomposing into soil 11 . Several studies had demonstrated the impact of straw returning on soil physicochemical properties, the structure of soil microbial community, and microbial activity [12][13][14] . Straw returning changed soil nutrient contents and enhanced the rice yield [15][16][17][18] . Yan et al 19 showed that rice straw returning signi cantly increased total organic carbon content, active soil organic carbon fractions, and soil microbial richness but did not affect the soil microbial diversity in Northeast China. In addition, Bu et al 20 found that straw returning had a dominant effect on bacterial community composition in a 12-year ricerice-rape rotation system. Studies had also shown that straw returning had signi cant effects in improving the activity levels of soil enzymes 21-23 .
Wu et al 24 showed that the straw returning increased the levels of urease, phosphatase, and invertase activities over a 5-year period. Zhang et al 25 observed that the fertilizer plus residues treatment led to higher potential activities of β-glucosidase, lignin peroxidase, and manganese peroxidase enzymes, whereas it reduced the activities of laccase enzymes in a 10-year fertilization experiment.
Compared with the conventional straw returning, the rice straw under the rice-cray sh integrated system was returned to the eld in the condition of long-term ooding, which can facilitate the cultivation of cray sh. In the rice-cray sh integrated system, the cray sh can feed on plant debris, and relate microbe produced during the ooded decomposition of straw. Meanwhile, the rapid decomposition of straw can be promoted by the excrement produced by cray sh during its growth. Therefore, these above phenomena may change soil physicochemical properties and microbial diversity.
Nowadays, studies related to rice straw returning in China were primarily based in rice cropping rotation or double rice cropping 18,26-27 . Some studies also focused on effect of gas emissions in the integrated rice-cray sh farming system, such as CH 4 , N 2 O, and NH 4 28-29 . However, there were few reports on the effects of straw returning on soil physicochemical properties and microbial diversity under the rice-cray sh integrated system. The aims of this paper were (a) to examine the changes of soil physicochemical properties, reducing substances properties, enzyme activity, and microbial community structure in the rice-cray sh integrated system and (b) to analyze the relationships between these changes under straw returning to provide reference for the exploration of improving soil fertility in the rice-cray sh integrated system and deliver theoretical guidance for maintaining healthy and sustainable agricultural ecosystem.

Soil physicochemical properties analysis
As shown in Table 1 treatment than that in the WS treatment. Meanwhile, the AN and the total N (TN) contents of the 10-20 cm layer were signi cantly higher in the WSC treatment than those in the WS treatment.

Soil reducing substances properties analysis
Soil total reducing substances, Fe 2+ , and Mn 2+ contents increased with increasing soil depth ( Table 2). Soil Fe 2+ content in the 0-20 cm layer, Soil Mn 2+ content in the 0-10 cm layer, and total reducing substances in the 0-20 cm layer was signi cantly lower in the WSC treatment than those in the WS treatment (P < 0.05). Meantime, the soil total reducing substances of the 0-10 cm layer was signi cantly higher in the WS treatment than that in the W treatment by 28.5% (P < 0.05). Soil enzyme activity analysis Soil enzyme activity decreased with increasing soil depth (Table 3). Soil sucrase and acid phosphatase activities showed an increasing trend in the WSC and WS treatments compared with the W treatment in all of the layers examined. Soil sucrase activity was signi cantly higher in the WS treatment than that in the W treatment (P < 0.05). Meantime, soil sucrase and acid phosphatase activities in the 10-20 cm layer were higher in the WSC treatment than those in the WS treatment by 44.4% and 32.2%, respectively. However, there was no signi cant difference in soil urease activity among different treatments in 0-20 cm layer. Soil functional diversity of microbial community analysis The soil average well color development (AWCD) value is one of the indices to judge the total carbon utilization ability of microbial community, re ecting the soil microbial activity and the diversity of physiological functions of microbial community 30 . With the increasing duration of culture ( Figure 1), the utilization degree of different carbon source increased gradually. In the 0-20 cm layer, the slope of the AWCD value was the highest at 24-120 h of culture, indicating that carbon source metabolic activity of soil microbes was the highest at this stage, and then a stable stage period was entered. During hours 72-192 in Fig. 1, the soil AWCD value in the WS treatment was higher than that in the W treatment, and the AWCD value of the WSC treatment was higher than that in the WS treatment in the 0-20 cm layer. This showed that the WSC and WS treatments increased microbial activity in the 0-20 cm layer.
The Shannon index re ects the diversity of bacterial community, with a higher index indicating the stronger the diversity of bacterial community [31][32] . The Simpson index re ects the changes in the population of each species, with a higher index indicating the position of dominant species more prominent 33 . The McIntosh index is an index to measure the community species consistency, with an increased degree of carbon source utilization, and consequently an increased McIntosh index 34 . Table 4 showed that at 96 h of culture in the Biolog ECO microplate, those indices were signi cant difference in the AWCD value in the 0-20 cm layer among the W, WS, and WSC treatments. In the 0-10 cm layer, the Shannon, Simpson, and McIntosh indices in the WS treatment were signi cantly higher than those in the W treatment. However, there were no signi cant difference in Shannon, Simpson, and McIntosh indices of the soil microbial community between the WSC and WS treatments. In the 10-20 cm layer, the Simpson index in the WS treatment had a signi cant increased compared with the W treatment. The Simpson, and McIntosh indices in the WSC treatment were signi cantly higher than those in the WS treatment. Therefore, the WS treatment increased signi cantly the diversity and species abundance of bacterial community in the 0-10 cm layer, the ability of utilizing carbon sources in the 0-10 cm layer, and the species abundance of bacterial community in the 10-20 cm layer. The species abundance of bacterial community, and the ability of utilizing carbon sources were increased signi cantly in the WSC treatment in the 10-20 cm layer. Analysis of soil microbe carbon source utilization Based on the types of carbon sources in the Biolog ECO microplate, the 31 kinds of carbon sources were divided into carbohydrates, carboxylic acids, amino acids, phenols, amines and polymers. The changes in the AWCD values of microbial carbon source utilization at 96 h of three treatments were analyzed.
As shown in Fig. 2, with increasing soil layer depth, the soil microbial utilization rate of carbohydrates, carboxylic acids, amino acids, phenols, amines, and polymers showed a gradual decreasing trend among three treatments. Compared with the W treatment, there were a huge increase in the WS treatment among the utilization rates of carbohydrates, carboxylic acids, amino acids, amines and polymers in the 0-20 cm layers by 47.5-52.4%, 31.4-63.0%, 27.9-45.7%, 1.4-51.8%, and 51.8-65.0%, respectively. The WSC treatment also represented the same tendency in the 0-20 cm layers compared with the WS treatment, in which the utilization rates of phenols in the 0-10 cm layer and the utilization rates of carbohydrates, amino acids, and amines in the 10-20 cm layer had signi cant differences. Consequently, the straw returning signi cantly increased amino acids and polymers utilization rates in the 0-10 cm layer, carboxylic acids utilization rates in the 10-20 cm layer. The straw returning + cray sh farming signi cantly increased the utilization rates of phenols in the 0-10 cm layer and the utilization rates of carbohydrates, amino acids, and amines in the 10-20 cm layer.
Interaction of physicochemical and reducing substances properties with soil enzyme activity and microbial diversity indices As shown in

Discussion
Some studies have reported that the straw returning and the rice-cray sh integrated system increased the concentrations of elements such as N, P, and K [35][36] . In the present study, the physicochemical properties were no signi cant differences between the WS and the W treatments in the 0-20 cm layers. A possible reason was that the time of the straw returning was so short that no su cient decomposition of the rice straw under waterlogging condition. Compared with the WS treatment, the WSC treatment had signi cantly higher in the AN and TN contents, supporting the ndings of Yi et al 37 . Meanwhile, the soil pH value decreased signi cantly under the WSC treatment in the 0-20 cm layers compared with the WS treatment. The reason for the decreased soil pH value may be the uneaten cray sh feed, the molts, excretions, and other substances produced by cray sh during their growth process, and dying cray sh, which were fermented and decomposed by bacteria under anaerobic conditions, producing a large number of organic acids and reducing soil pH.
Soil enzyme activity is often considered an important indicator of soil quality. Urease, acid phosphatase, and sucrase were related to the N, P, and C cycles, and their activities showed an increasing trend in the WS treatment compared with the W treatment in all of the soil layers examined. Sucrase activity was signi cantly higher in the WS treatment than that in the W treatment in the 0-20 cm layers. Our results were in agreement with those of Wu et al 24 . Meanwhile, compared with the WS treatment, acid phosphatase and sucrase activities were signi cantly higher in the WSC treatment in the 10-20 cm layer. A more important factor was the incorporation of a mass of straw to the upper soils, which can increase the microbial population 24 . Another factor was the cray sh digging and other activities which disturbed the surface of paddy soils and increased soil permeability, enhanced the soil nutrients and dissolved oxygen in water to reach the deep layer, thus changing the microbial activities.
Soil reducing substance is an important indicator of soil redox status, which has a signi cant impact on crop growth and yield 38 . Our study showed that the Fe 2+ , Mn 2+ contents, and total reducing substances were prominently lower in the WSC treatment than those in the WS treatment after longterm waterlogging. Our results were difference with some studies 4,39 . The chief factor was the difference in the pilot areas. In our study, the experimental site belonged to typical waterlogged paddy in the Jianghan Plain, having strong soil reduction, and slowing organic matter decomposition, thus leading to soil gleying. However, cray sh penetrated the surface and base layers of the rice paddy soil, increasing water migration channels, and allowing the nutrients and dissolved oxygen in water to reach the base layer, thus decreasing the soil reducing substance contents here. Meanwhile, in our study, the straw returning decreased signi cantly the total reducing substances in the 0-10 cm layer. The incorporation of a mass of straw to the upper soils may be the reason, which can further deepen the anaerobic environment of the soil.
Soil microbial community functional diversity is an indicator of soil microbe community structure and function, and re ects the ecological characteristics of soil microbes 40 . The Biolog ECO method is a sensitive method for detecting functional changes in soil microbial community structure, and is broadly used to assess soil microbial community functional diversity 41 . In this study, we observed that the AWCD value of the WS and the WSC treatments always higher than those in the W treatment during all cultured time. Meanwhile, the WS treatment increased the soil microbial diversity, soil microbial richness, and the ability of utilizing carbon sources in the 0-20 cm layers. Those were in good agreement with those of Yu et al 42 . The reason for this was the straw returning contributed to the greater accumulation of rice residues on the soil surface, promoting the growth and proliferation of certain groups of microbes. Compared to the WS treatment, the WSC treatment increased signi cantly the microbial utilization rate of carbon source and the species abundance of bacterial community in the 10-20 cm layer. Those were consistent with the results reported by Zhu et al 43 . It may due to the reason that the cray sh activity could enhance the dissolved oxygen in water to reach the deep layer, change the availability of nutrients and oxygen, and affect microbial community structure, function, and diversity.
Research on the utilization of the different carbon sources by soil microbe could reveal detailed information about microbial community metabolism 44 . In this study, the results showed that the straw returning and cray sh farming had a greatly in uence on the community composition of soil microbe in the 0-20 cm layers. The straw returning can increased the microbial populations and the cray sh farming alleviated the anaerobic situation of soil and increased aerobic bacterial community, which made the utilization of carbon sources for microbe increase at this time.
Soil microbial diversity was affected by soil type, moisture, pH, and soil management measures 45 . The correlation analysis showed that the soil pH value, AK, TN, and Fe 2+ contents and total reducing substances were signi cantly correlated with microbial diversity indices. Meanwhile, compared with the other two enzymes, acid phosphatase was more affected by soil physicochemical and reducing substances properties.

Study area and experimental design
The study was conducted on a 7-year-old rice-cray sh integrated system in waterlogged paddy elds at Immigrant village, Houhu Farm, Qianjiang City, Hubei Province, China (112°41′32.5″E, 30°22′41.2″N). This area was part of the low lake areas of Jianghan plain, with the static groundwater level of 40-60 cm in winter, belonging to the humid climate zone of north subtropical monsoon. The average annual temperature was 16.1°C, with a frost-free period of 246 days. The average annual rainfall was 1100 mm. The soil type was a uvo-aquic paddy soil, which developed from lacustrine deposit.
The eld experiment started in 2014, and three treatments were set up: winter ooded fallow + no straw returning (W), winter ooded fallow + straw returning (WS) and winter ooded fallow + straw returning + cray sh farming (WSC). Each treatment was set to be repeated 3 times, with a cell area of 100 m 2 .
The rice cultivar of three treatments was Jianzhen 2. The use of plant in the present study complies with international, national and/or institutional guidelines.
The way of the rice straw returning was high stubble followed by irrigation and rotary tillage before rice planting. The stubble height was about 40 cm and the return amount was 1900 kg·ha -1 .
Cray sh larvae (weighing 5 ± 2g) were stocked at a density of 9× 10 4 larvae·ha −1 , and the cray sh self-propagated inside the rice paddies. Then a proper amount of broodstock were added at this time of the year according to the actual situation. The cray sh were released into the ooded eld on 25 October 2014, exactly 20 days after the rice harvest. Mature cray sh were harvested in June in the second year, and immature cray sh migrated to the peripheral trench before re-entering the rice eld after eld puddling, seedling planting, eld drying, and re-watering. In the second season, mature cray sh were harvested before the rice harvest. In order to prevent cray sh from escaping, ditches with a width of 0.4 m and a depth of 1.0 m were set in each community, while cray sh ditches with a width of 3.0-4.0 m and a depth of 0.8-1.0 m were set at one side of the community, and nylon nets were also set around it.
Before the experiment in 2014, the basic physical and chemical properties of 0-20 cm topsoil were as follows: pH 7.12, total organic carbon (TOC)

Soil sampling and storage
Soil samples were collected on the 11 th November 2020 after the harvest using a sample auger at depths of 0-10 cm and 10-20 cm. Sampling was conducted from ve different sites within each replicate plot, and the ve samples were mixed to prepare a composite sample for the plot. Immediately after sampling, visible root fragments and stones were manually removed, and the samples were mixed well and divided into two portions. One portion of fresh soil was passed through a 2-mm sieve and stored in a refrigerator at 4°C until its biological characteristics were analysed, and the other was air-dried and ltered in preparation for chemical characteristics analysis.

Soil physicochemical properties analysis
Soil physicochemical properties were assayed according to the methods described by Bao 46 . Soil pH was measured in a soil water mixture (1:2.5w/v) using a pH meter. The TOC content was determined by oxidation with potassium dichromate and titration with ferrous ammonium sulfate. The TN was determined by the Kjeldahl digestion method. The TP and TK were extracted and determined by the perchloric acid digestion methods and spectrophotometer protocols. The soil AN was converted to NH 4 + under alkaline conditions, collected in H 3 BO 3 solution, and then determined by titration with standard 0.01 mol·L -1 H 2 SO 4 . The AP was determined using the molybdenum blue method with a spectrophotometer after extraction with 0.5 mol·L -1 NaHCO 3 at pH 8.5. The AK was determined by 1.0 mol·L -1 ammonium acetate extraction-ame spectrophotometry.

Soil reducing substances properties analysis
The potassium periodate colorimetry was used for Mn 2+ , phenanthroline colorimetry was used for Fe 2+ , and Al 2 (SO 4 ) 3 leaching-potassium dichromate volumetric method was used for the total amount of reducing substances 47 .

Soil enzyme activity analysis
Enzyme activity was assayed according to the methods described by Guan 48 , and acid phosphatase activity was estimated by determining the amount of phenol released after incubating the samples with phenyl disodium phosphate (0.5% w/v) for 24 h at 37°C. Urease activity was measured by determining the amount of NH 4 + released from a hydrolysis reaction after incubating the samples with urea (10% w/v) for 24 h at 37 °C, and sucrase activity was measured by determining the amount of glucose released after incubating the samples with sucrose (8% w/v) at 37°C for 24 h.
Soil functional diversity of microbial community analysis Soil functional diversity of microbial community was determined by the Biolog-ECO method 49 . The 10 g fresh soil was passed through a 2 mm sieve, added to 100 mL 0.85% (w/v) sterile NaCl solution, and agitated for 30 mins. In the situation of bacteria free, the mixture was diluted to 10 −3 with sterile 0.85% NaCl solution, and an 8-channel pipette was used to add 150 μL of the diluted suspension to each well of the Biolog ECO plate. Each soil sample was done in triplicate. Cultures were grown at a constant temperature of 25°C. The absorbance at 590 nm was measured for each well after 24, 48, 72, 96, 120, 144, 168, 196, and 240 h.
(1) Average absorbance (AWCD) can judge the total ability of microbial community to utilize carbon source.
In this equation, A i is the relative absorbance at 590 nm of the ith reaction well; A A1 is the relative absorbance of well A1; wells with an A i -A A1 value of <0 are set to 0 for the purpose of calculations, so that all values of A i -A A1 are greater than or equal to 0.
(2) Shannon index H' is used to evaluate richness.
Here, P i is the ratio of the relative absorbance of the ith well to the total relative absorbance of the entire sample plate.
(3) Simpson index is used to evaluate dominance index, and its variant Gini index is often used to evaluate diversity.
Here, P i is the ratio of the relative absorbance of the ith well to the total relative absorbance of the entire sample plate.
(4) McIntosh index is a diversity index based on multi-dimensional spatial distance of community species. In fact, It is a measure of consistency.
Here, n i is the relative absorbance of the ith well.

Statistical analysis
The data were analysed using Excel 2010, SPSS software (version 22.0), and the treatment means were compared using the least signi cant difference test at P<0.05. Pearson's correlation analyses were conducted to investigate relationships between soil physicochemical properties and biological parameters. All experimental protocols involving plant materials were conducted in accordance with institutional, national, and international guidelines and legislation.

Data Availability
The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request. Status of different carbon sources utilized by soil microbes. Means with different letters for the same property in the same soil layer indicate signi cant differences at P < 0.05.