3.1. Rice–Crayfish Integrated Farming Modes
Rice–crayfish integrated farming can be divided into two modes: rice–crayfish rotation (RCR) and rice–crayfish mutualism (RCM); the spatio-temporal characteristics and irrigation management of the two modes are displayed in
Figure 3. During the period from rice harvest to new season transplanting, the paddy field is used for crayfish culture with long-time deep water irrigation (30–50 cm), which is consistent in both RCR and RCM. The pivotal difference between the two modes is whether semi-deep irrigation (20–40 cm) is conducted to induce crayfish into the paddy field during the rice growth process as RCR does not do this, whereas RCM does.
Among the surveyed farmers, 76.9% and 23.1% undertook RCR and RCM, respectively (
Table 1). For farmers selecting RCR, most of them considered this mode simple and convenient (84.4%), particularly for beginners with limited experience, since the rice planting and crayfish culture are independent from each other in RCR. Nearly two thirds of the surveyed farmers thought that they could obtain a steady rice yield (66.9%) as the rice cultivation is quite similar to that in rice monoculture and avoid extensive management of crayfish in July and August when it is difficult to catch crayfish due to hot weather, requiring more fodder and labor supply. Meanwhile, the water supply (60.6%) and fodder cost (63.1%) were thought to be less required by more than 60% of farmers who selected RCR, and 29.4% of surveyed farmers choosing RCR were partly attracted by the great profit potential of this mode due to its higher crayfish yield potential; they believed that they could dedicate more time to harvest crayfish before rice transplanting. Only 8.1% of the surveyed farmers considered the higher rice yield potential as a reason for selecting the RCR mode. For farmers selecting RCM, they mainly preferred the advantage of culturing crayfish with a bigger size (60.4%) and larger quantity (83.3%), while 41.9% of the surveyed farmers were convinced that RCM had the advantage of promoting excellent-quality rice growth because of the lower pesticide application; they reported that crayfish could eat parts of pests in the paddy field. About one fifth (20.8%) considered the higher rice yield potential as a reason for their selection of RCM; they believed that they could cultivate rice varieties with a longer growth duration since the time between crayfish harvest and rice planting was not very pressing in the RCM mode. In 2020, we carried out a return visit to some of the farmers we reviewed last year, finding that a few of them would like to switch from RCR to RCM as they gradually realized the beneficial impact of RCM on bigger-size crayfish formation and pest and weed control.
Based on face-to-face conversations, we further confirmed that many farmers did not have a profound understanding of the advantages and disadvantages of the two RCIF modes and thus did not know how to select a suitable mode for themselves. For beginners and farmers with fund shortages, we recommended RCR to be a more convenient mode, as per the reasons mentioned above. Additionally, RCR requires less investment in field engineering than RCM since this mode requires less or even no ditches, whereas the RCM mode requires a larger area of ditches. For farmers with rich experience in crayfish culture, we suggested that they could try to conduct the RCM mode to strive for better profits through culturing second-season crayfish after rice transplanting. Farmers should have adequate recognition of each RCIF mode and their requirements for rice cultivation techniques. Then, they should determine an RCIF mode and production scale in consideration of their capital, knowledge, experience in aquaculture and the capacity for rice market exploitation.
The results showed that farmers commonly pay much more attention to crayfish than rice when selecting an RCIF mode due to their definite advantage in output value, and this tendentiousness had caused negative impacts on rice production security. We thought that this unbalanced profit model would be unsustainable once the price crayfish of dropped. To promote sustainable development of RCIF, we suggested that the position of rice cultivation should be strengthened to form a more balanced relationship between rice planting and crayfish culture.
3.2. Rice Yield
The two investigation years, 2019 and 2020, generally had normal climate conditions during the rice growth period; thus, the yield performance of the two years could reflect well the local current rice cultivation level in RCIF. In this survey, we directly obtained or estimated rice yield data with ditches from 126 farmers (
Table 2). The moisture content of grains was uniformly adjusted to 14.0% for calculating the rice yield. The statistical results indicated that 27.1% and 55.0% of surveyed farmers achieved rice yields of 6000–7500 kg·ha
−1 and 7500–9000 kg·ha
−1, respectively, jointly accounting for the major rice yield range in RCIF. In total, 8.6% of surveyed farmers obtained rice yields of 4500–6000 kg·ha
−1 and 9.3% of farmers obtained yields of 9000–10,500 kg·ha
−1. Survey results showed that over 35.9% farmers obtained rice yields less than 7500 kg·ha
−1 and failed to meet the request for rice yields in RAAIF, according to the document offering technical specifications for the integrated farming of rice and aquatic animals, which stipulates that the rice yield should be at least 7500 kg·ha
−1 in RAAIF production.
Based on field surveys, we detected that higher proportions of circumjacent ditches in paddy fields was a primary reason for the dissatisfactory levels of rice yield per unit. To ensure adequate rice planting area, the Ministry of Agriculture and Rural Affairs of the People’s Republic of China stipulates that the proportion of circumjacent ditches should be less than 10% in field engineering for RAAIF, but there still exists some phenomenon of excessive field engineering in local RCIF, which is driven by farmers’ intense yearning for higher crayfish yields. Local related departments ought to enhance supervision of paddy field use to avoid a decrease in actual rice production area. Besides excessive field engineering, rice culture techniques were also significantly responsible for the current rice yield levels; we therefore investigated other parameters related to rice production. The results are presented in the following part of this paper.
3.3. Selection of Rice Varieties
Local farmers used to plant indica rice as they had a preference towards a shorter growth duration and less fertilizer and water requirements. As displayed in
Table 3, 78.4% and 21.6% of the surveyed farmers used indica rice and japonica rice, respectively, illustrating that indica rice still occupies the dominant position in this integrated farming system. In fact, the indica rice and japonica rice used by farmers in RCIF were almost hybrid indica rice and conventional japonica rice, respectively.
In this survey, a shorter growth duration was also emphasized by farmers in RCIF, being reported by 72.4% of farmers planting indica rice (RPIR) and 62.2% of farmers planting japonica rice (RPJR). Evidently, a high yield potential (79.1% of PRIP, 77.8% of RPJR), resistance to lodging (82.8% of PRIR, 75.6% of RPJR) and resistance to pests and diseases (90.2% of PRIR and 82.2% of PRJR), and shorter growth duration were major reasons for farmers to select a specific cultivar, reflecting that most farmers appreciated all-sided characteristics when choosing a rice variety. It should be pointed out that semi-deep irrigation in the RCM mode would increase the rice lodging risk, based on our recent study [
12], so RCM would require more lodging-resistant rice varieties than RCR. It was a consensus that a shorter rice growth duration in paddy fields could be beneficial to crayfish culture for both RCR and RCM as a shorter growth duration of rice could provide more time for crayfish to survive in the paddy field, which could promote its higher yield and bigger size formation, and early harvesting of rice could advance the time to market of the next year so as to acquire a higher unit price of crayfish. All of these are beneficial for farmers in obtaining more crayfish income, and RCR presents a more intense request for shorter rice growth duration as this mode only harvests crayfish once.
More than one third of farmers selecting indica rice (38.0%) regarded the low seed price as a vital factor for the rice variety selection, while the same reason was reported by 15.6% of farmers selecting japonica rice; this difference in the recognition degree of the seed price was probably due to the difference in the unit price, in which hybrid indica rice is almost tenfold that of conventional japonica rice. For farmers choosing japonica rice, about sixty percent emphasized the rice quality because rice with excellent quality would more easily be accepted by rice milling companies with a higher price. Meanwhile, in order to obtain higher profits, a few farmers would like to process and pack rice in cooperation with factories and then sell it through stores or e-commerce. In total, 22.7% and 20.0% of the surveyed farmers emphasized the rice’s resistance to high temperatures since they used to sow early under RCIF, which may make rice encounter high temperatures more frequently during the reproductive stage. Also, 9.8% of the surveyed farmers reported that rice plants with taller heights were useful for the RCM mode due to its shading effect against water and soil in the paddy field, which could provide a more comfortable environment for crayfish growth.
The survey results reflected that RCIF’s general requirement for the rice variety were distinguished from those in rice monoculture and discrepant demand also existed between RCR and RCM. Although local seed distributors publicized advantages of some rice varieties’ fitness for RCIF, there were indeed few rice varieties that could meet the demands faultlessly for RCIF currently, according to the results of a three-year trial planting by our studying team and reports from local farmers and the agricultural extension department. Hence, suitable rice varieties for each RCIF mode should be developed to adequately meet the requirement of rice culture in RCIF.
Unlike the diversity of rice varieties, crayfish breeding was just at primary stage in China, there were no formal crayfish variety by far. Crayfish seed was universally purchased from breeding ground nearby or retained by farmers themselves, thus this study did not concern crayfish variety.
3.4. Rice Planting Pattern
This survey revealed that there were four planting patterns in RCIF, including manual transplanting, mechanized transplanting, manual direct seeding, and mechanized direct seeding. They accounted for proportions of 57.7%, 21.6%, 1.4%, and 19.2%, respectively (
Figure 4). Farmers using manual transplanting believed that this planting pattern can discretionarily adjust their transplanting data due to its elastic seedling age, which could well adapt to the frequent occurrence of transplanting with long seeding ages in RCIF. Manual transplanting’s advantage in elastic seedling age could help farmers gain more days in crayfish culture and harvest before rice transplanting. Although mechanized transplanting was quite popular in local RWR production, it occupied less than a quarter (21.6%) of RCIF. According to the claims by surveyed farmers, there were two main reasons responsible for their unwillingness to use mechanized transplanting; one was that they would like to end crayfish harvesting at an uncertain date, which was rather contradictory to the tense seedling age of mechanized transplanting, and another dominant reason was that rice transplanters usually work inefficiently and had a high risk of becoming trapped in sticky rotten soil, which is caused by long-time deep or semi-deep irrigation in crayfish culture. The low mechanized transplanting rate in RCIF is unsustainable with the continuous reduction of the rural labor force; there is an urgent request for the development of mechanized rice planting. To cope with the difficulty of mechanized transplanting in RCIF, we recommend that three key techniques ought to be developed to fit special working environments under RCIF. Firstly, farmers should take measures to loosen soil to decrease soil resistance, such as organic fertilizer application and appropriate deep tillage; secondly, agricultural machinery companies should develop new rice transplanters with larger-diameter wheels with better friction force and more powerful engines so as to ensure steady transplanting efficiency under high-resistance working environments. Thirdly, farmers should maintain the rice seedling height between 20 and 25 cm in the transplanting stage using plant growth regulators, especially for long seeding ages, with the aim to make transplanters work more easily in the paddy field. Direct seeding was less adopted by farmers because slather herbicide had to be applied to control weeds, which would pose a great threat to crayfish survival. Additionally, a worse rice quality also discouraged farmers from direct seeding, especially for farmers who expected to attain a higher income by selling premium-quality rice.
In addition, we deemed that the rice planting pattern has a close connection with the RCIF mode. If farmers select the RCM mode, they ought to use cultivation techniques that could promote rice seedling recovery and tiller occurrence quickly. Therefore, direct seeding is not appropriate for the RCM mode and the transplanting pattern should be improved to decrease the injury of seedlings during the transplanting process.
3.5. Seeding Period
In total, 79.3% of the surveyed farmers selected a transplanting pattern and their seeding period ranged from early April to early June (
Table 4). There was only 8.5% of surveyed farmers who chose to seed in April, and they mostly transplanted rice in early and middle May. Through face-to-face conversation, we found that this group mainly sold crayfish seed rather than mature crayfish so that they could transplant rice once the crayfish are adequately harvested. The survey results showed that nearly four fifths of the surveyed farmers conducted seeding in May, especially in late May (38.0%) and middle May (27.9%), and they accordingly transplanted rice mainly in June or late May. However, 8.5% of the surveyed farmers selected to undertake seeding in early June, which was a seeding time that rarely occurred in local RWR except for the direct seeding pattern. Based on the information provided by this group of surveyed farmers, they postponed the seeding time in consideration of gaining more time for harvesting crayfish before transplanting. Indeed, many farmers would like to delay the transplanting time to obtain more days to culture crayfish, in accordance with face-to-face conversations. In theory, there were two methods of achieving late transplanting: one is delaying the seeding time and the other is transplanting rice seedlings of older ages. For the former method, the most popular rice varieties currently would be mature until November if seeding were delayed to June and transplanting was delayed to early July, which would present a negative impact on the next year’s crayfish culture. For the latter method, if seeding was carried out at a proper time and seedlings were transplanted at more than 25 days of age, farmers tended to conduct dry-raised seedling and transplant seedling manually to cope with the longer seedling age and tall seedling height, since it is difficult for current transplanters to transplant larger rice seedlings over 25 days of age. However, farmers had to face the question of labor shortages and high labor costs in the transplanting stage if transplanting long-age seedlings, which was popular in Xuyi county due to the rural labor transfer during the process of urbanization. To meet the request for delayed transplanting, a shorter growth duration variety needs to be developed and the mechanical transplanting pattern should be adjusted to adapt to seedlings of at least 30 days seedling-age and at least 20 cm height.
Results showed that 20.6% of farmers used direct-seeding. According to face-to-face conversations, farmers selecting direct-seeding preferred its convenient operability and shorter rice growth duration, and usually did not expect a higher income from rice. Direct-seeding in June was reported by 86.0% of surveyed farmers, with middle June ranking the highest (44.2%), since farmers could coordinate the rice and crayfish growth durations by seeding in this period. However, 9.3% and 4.7% of the surveyed farmers chose to seed in May and early June, respectively. According to our face-to-face conversations, farmers who implemented direct-seeding in May majorly emphasized selling crayfish seed and ended crayfish culture much earlier than farmers who emphasized culturing bigger-size crayfish. Two of the farmers surveyed conducted direct-seeding in early June based on their use of a quite shorter growth-duration hybrid indica rice variety so as to obtain more days for crayfish culture in the paddy field.
3.6. Nitrogen Nutrient Management
In local RWR production, the recommended nitrogen application levels for indica rice and japonica rice in Xuyi county generally range from 180 to 225 kg·ha
−1 and 270 to 300 kg·ha
−1. This survey investigated the nitrogen level applied by farmers in RCIF (
Table 5). In total, 25.2% 42.9%, 29.4%, and 2.5% of the surveyed farmers planting indica rice applied nitrogen in amounts of <135 kg·ha
−1, 135–180 kg·ha
−1, 180–225 kg·ha
−1, and >225 kg·ha
−1, respectively. For farmers planting japonica rice, 17.7%, 40.0%, 35.6%, and 6.7% applied amounts of N of <180 kg·ha
−1, 180–225 kg·ha
−1, 225–270 kg·ha
−1, and >270 kg·ha
−1, respectively, illustrating that RCIF reduced the N application amount greatly when compared with RWR. Evidently, there are some possible factors affecting soil characteristics in RCIF compared with rice monoculture, including (but not limited to) fodder application, waterweed returning, crayfish excretion, and burrowing behavior. Many sources in the literature reported that culturing aquatic animals in paddy lead to increased soil fertility and microbial communities [
10,
14,
15,
16,
17,
18] and that this effect would be more significant with the increase in culture time. But, soil secondary gleization was observed after continuous RCIF [
19], revealing that RCIF may have a two-sided influence on soil quality. However, there are less studies concerning the influence of RCIF, especially RCM, on the physicochemical properties of rice soil by far, let alone corresponding nutrient management strategies. Therefore, efforts ought to be made to research the change in soil characteristics due to RCIF to adjust fertilizer application schemes correspondingly.
3.7. Pest and Disease Occurrence and Their Management
Eight pests and diseases were recorded from surveyed farmers’ claims (
Table 6). For indica rice, bacterial leaf streak was the most incidental disease, being reported by 40.5% of the surveyed farmers. Other pests and diseases ordered from high to low were rice false smut (22.7%), rice sheath blight (11.7%), rice bacterial foot rot (9.8%), rice borer (8.0%), rice blast (6.1%), rice leaf roller (3.1%), and rice hopper (1.8%). For japonica rice, rice false smut ranked the highest proportion (28.9%), and rice blast, rice sheath blight, and rice borer were, respectively, reported by 17.8%, 15.6%, and 11.1% of the surveyed farmers, while 8.9%, 6.7%, and 2.2% of the surveyed farmers claimed the occurrence of rice bacterial foot root, rice leaf roller, and rice hopper in their practices.
In recent years, rice bacterial leaf streak has occurred frequently during the middle and late growth stages for indica rice under both rice monoculture and RCIF in the Hongze Lake district, presenting stress on rice growth; this disease is difficult to control once it happens [
20]. Until now, only a small number of local indica rice varieties had strong rice bacterial leaf streak resistance, based on claims from the local agricultural extension department. To restrain rice bacterial leaf streak, there is a necessity to undertake seed treatment with a specific bactericide before seeding, but this step is usually ignored in RCIF. Additionally, isolated irrigation for each field could decrease the possibility of rice bacterial leaf streak spread. Moreover, rice bacterial leaf streak resistance genes should be determined and applied in rice breeding in the future. Rice false smut had a frequent incidence rate for both indica rice and japonica rice, which may be partly due to the high relative humidity caused by circumjacent ditches, especially in the semi-deep irrigation of the RCM mode.
On the whole, diseases exhibited a significantly higher occurrence than pests for rice in RCIF. We investigated farmers who engaged in RWR formerly to compare the occurrence of rice pests and diseases between RCIF and RWR. Most of them hold the opinion that pests could be reduced to some extent more in RCIF than RWR, while there was no positive influence of RCIF on preventing or controlling diseases and rice false smut even presented a higher occurrence frequency in RCIF. Previous studies have analyzed the effects of aquatic animals on rice pests and diseases in RAAIF systems, and it has been reported that some pests and diseases could be controlled well in some RAAIF systems [
21,
22,
23]. For instance, researchers have observed a scenario in which fish hit the rice stem and ate dropped insects under rice–fish mutualism [
2]. Compared with other aquatic animals, crayfish is a relatively new species in paddy fields, so its impact on rice pests and diseases has not been systematically studied. Currently, efforts should be made to understand the regularity of the occurrence of rice pests and diseases in RCIF and their relationship with crayfish culture.
We also investigated control measures for rice pests and diseases in RCIF (
Table 7). In total, 29.8% of surveyed farmers use chemical pesticides to control weeds, pests, and diseases, and only 7.7% of the surveyed farmers selected to control pests and diseases without chemical pesticides. They may instead use biological pesticides and other measures, such as planting vetiver grass beside the paddy field, which a plant that could be applied to deal with borers due to its volatile nature [
24].
The rest, 62.5%, of the farmers used both chemical pesticides and non-chemical measures to control pests and diseases. In general, rice production under RCIF still relies on chemical pesticides for pest and disease control presently. Some chemical pesticides with higher toxicities are not encouraged for application in RCIF by agricultural departments, so farmers usually built ridges between the paddy field and ditches to avoid the spreading of pesticides into the ditches, which crayfish mainly inhabit after rice transplanting in the RCR mode. For RCM, chemical pesticides were strictly limited in their application to guarantee the crayfishes’ security as rice and crayfish live together in the paddy field. No surveyed farmer reported incidents where crayfish were harmed by chemical pesticides, this may benefite from farmers’ appreciation of field engineering and their cautious application of chemical pesticides. In recent years, weed control usually needed more pesticides than pests and diseases in the Hongze lake district. Under the RCM mode, some types of weed occurrence could be suppressed by flooding irrigation and the feeding effect of crayfishes, but there were still many types of weeds that could survival under flooding, even with the semi-deep irrigation of RCM, such as monarch redstem and monochorea. To deal with flood-tolerant weeds, farmers had to employ labor force to manually remove the weeds, which significantly increased the cost of RCIF; thus, efficient and crayfish-safe pesticides are required to be developed to control weeds. In rice production, the specific biology could be used to attract target pests by its own alluring substance and kills them, which is called biological trapping. Our investigation results indicated that physical and biological trapping was less used by farmers. We recommend that biological trapping and physical control measures be appreciated and generalized in the future.
3.8. Main Rice Culture Problems in RCIF
As shown in
Table 8, there were six rice culture constraints reported by more than half of the surveyed farmers, including weak recognition of requirements of rice culture for a specific RCIF mode (73.6%), lack of suitable rice varieties (77.9%), lack of suitable agricultural machinery for RCIF (81.7%), shortage of pesticides resistant to pests and diseases but are healthy to crayfish (76.4%), and it is hard to determine a N fertilizer application scheme (74.5%), indicating that these problems were quite popular in RCIF. They were generally similar to our analysis in the above contents of this paper. Another constraint offered by a few farmers who conducted RCM was that they were unaware of the proper period for rice–crayfish mutualism with deep irrigation (20.2%). It has been widely recognized that water management for high-yield rice cultivation is mainly based on shallow and wet irrigation and drainage for controlling ineffective tillers is necessary [
25,
26,
27]. Although semi-deep irrigation is beneficial to crayfish culture, it has also shown an adverse impact on rice lodging resistance, according our recent research [
12]. Therefore, it is necessary to understand that when to undertake semi-deep irrigation to both promote crayfish growth and to reduce the negative impact on rice development. However, related information remains unclear.
From these common constraints, we detected that farmers had limited knowledge of appropriate rice cultivation techniques under RCIF currently. To our knowledge, there has been a distinct lack of information about the impact of RCIF on the rice growth environment in various aspects, such as changes in the soil physicochemical properties under continuous feeding, Elodea nuttallii returning, occurrence of pests and diseases, field microclimate, performance of rice lodging resistance, and others, which could have significant effects on the rice yield and quality. Only based on sufficient understanding of these questions could useful countermeasures be suggested to improve the rice cultivation level in RCIF. Consequently, it is quite urgent for researchers to pay more attention to the influence of RCIF on the rice growth environment from various aspects, like the soil characteristics, microclimate of the paddy field, and plant physiology.