Exploring the emergence and changing dynamics of a new integrated rice-crawfish farming system in China

Wei, Yanbing et al. Article — Published Version Exploring the emergence and changing dynamics of a new integrated rice-crawfish farming system in China Environmental Research Letters Provided in Cooperation with: Leibniz Institute of Agricultural Development in Transition Economies (IAMO), Halle (Saale) Suggested Citation: Wei, Yanbing et al. (2023) : Exploring the emergence and changing dynamics of a new integrated rice-crawfish farming system in China, Environmental Research Letters, ISSN 1748-9326, IOP Publishing, Bristol, Vol. 18, Iss. 6, pp. 1-13, https://doi.org/10.1088/1748-9326/acd8d2 , https://iopscience.iop.org/article/10.1088/1748-9326/acd8d2 This Version is available at: https://hdl.handle.net/10419/271664 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. http://creativecommons.org/licenses/by/4.0 Environ. Res. Lett. 18 (2023) 064040 https://doi.org/10.1088/1748-9326/acd8d2 OPEN ACCESS RECEIVED 14 March 2023 REVISED 3 May 2023 ACCEPTED FOR PUBLICATION 25 May 2023 PUBLISHED 6 June 2023 Original content from this work may be used under the terms of the Creative Commons Attribution 4.0 licence. Any further distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI. LETTER Exploring the emergence and changing dynamics of a new integrated rice-crawfish farming system in China Yanbing Wei1,2, Daniel Müller2,3,4, Zhanli Sun2, Miao Lu1, Huajun Tang1and Wenbin Wu1,∗ 1State Key Laboratory of Efficient Utilization of Arid and Semi-arid Arable Land in Northern China/Institute of Agricultural Resources and Regional Planning, Chinese Academy of Agricultural Sciences, Beijing 100081, People’s Republic of China 2Leibniz Institute of Agricultural Development in Transition Economies (IAMO), Halle (Saale) 06120, Germany 3Geography Department, Humboldt-Universität zu Berlin, Berlin 10099, Germany 4Integrative Research Institute on Transformations of Human-Environment Systems (IRI THESys), Humboldt Universität zu Berlin, Berlin 10099, Germany ∗Author to whom any correspondence should be addressed. E-mail: [email protected] Keywords: crop-aquaculture system, crawfish, paddy rice, spatio-temporal dynamics Supplementary material for this article is available online Abstract Crop-aquaculture systems are widely adopted around the world as they can provide high protein and energy outputs per unit of land and raise farm incomes, particularly for smallholder farmers. Recently, a new crop-aquaculture system, which combines rice production with crawfish breeding (integrated rice-crawfish farming), has emerged and rapidly expanded in China. However, the spatial extent and temporal dynamics of this integrated farming system largely remain unclear, which prohibits rigorous impact assessments to support its sustainable development. Here we use time series of Landsat satellite data, for the first time, to explore the emergence and the changing dynamics of this rice-crawfish farming system for the period of 2013–2021 in five provinces (805 600 km2) of China, where 90% of the global crawfish are produced. Our analysis reveals that the total area of rice-crawfish farming in these five provinces increased steadily from 0.11 Mha in 2013 to 0.70 Mha in 2019, then sharply contracted by a third in 2020 and rebounded in 2021. Spatially, rice-crawfish system is located primarily in low-elevation plain areas with abundant water resources, where paddy rice cultivation has traditionally dominated agriculture. More concentrated rice-crawfish distribution is observed in Jianghan Plain, and regions around Dongting Lake and Poyang Lake. The spatial distribution of rice-crawfish cultivation experienced considerable expansion towards the east and north from 2013 to 2021, with the largest expansion found in Jiangsu and Anhui after 2018. At the county level, over 6% of counties experienced notable area increases of more than 60 km2from 2017 and 2019, but 20% of counties have decreased from 2019 to 2021. Among the converted land use types, irrigated cropland is the largest contributor to rice-crawfish expansion with a contribution of 56%, followed by water bodies (25%) and rainfed cropland (13%). The spatial and temporal information provided in this study helps to understand the evolution of rice-crawfish cultivation in China and facilitates more efficient management of land resources under the rapid development of this farming system. 1. Introduction We are entering a stage which requires producing more diverse foods with less resource and environment effects (Devi et al 2017, Siddiqui et al 2021). In this regard, integrated food production systems such as crop-aquaculture systems are highly encouraged as they can provide high protein and energy outputs per unit of land and raise farm incomes (Nhan et al 2007, Singh and Singh 2017, Costello et al 2020, Bernhardt and O’Connor 2021). Aquaculture, the production of aquatic animals under controlled conditions, has reduced pressure on capture fisheries and is expected to surpass the volume of seafood from stagnant © 2023 The Author(s). Published by IOP Publishing Ltd Environ. Res. Lett. 18 (2023) 064040 Y Wei et al capture fisheries by 2030 (FAO 2020a, Cottrell et al 2021). Efforts are being made to maximize the efficiency of aquaculture as a food production system by increasing total yield and density (Stevens et al 2018, Naylor et al 2021, Thilsted 2021). One such promising resource recycling strategy is the integration of aquaculture and crop production, which has been practiced for thousands of years, particularly in East and South Asia (Xie et al 2011, Islam et al 2015). China is the global largest supplier of both rice and aquaculture products, accounting for 30% and 58% of global production, respectively (FAO 2019, 2020a). Considering the limited cropland resources in China, how to guarantee sufficient protein and caloric output on limited available land is a pivotal issue both for the government, to ensure domestic food security, and for farmers, who aim to generate competitive incomes. Co-cultivation of crops and aquaculture farming systems are well-developed practices in China and can make more efficient use of resources relative to traditional cropping patterns (Ahmed and Garnett 2011). Red swamp crawfish (Procambarus clarkii), native to northeastern Mexico and the southern USA, have recently become a flourishing commodity in China due to skyrocketing demand for crawfish as a popular street and snack food (Zhang et al 2021). Crawfish become one among the most popular freshwater species for aquaculture in China (Jin et al 2019). In 2019, domestic production of crawfish reached two million tons, accounting for 96% of global production (FAO 2020b). Approximately 86% of all crawfish in China are bred in paddy fields (National Bureau of Statistics of China (NBSC) 2020a). Rice-crawfish farming system has a much higher pure profit comparing with traditional rice cultivation, with 23 ×103Yuan ha−1compared to 6 ×103Yuan ha−1(Hou et al 2020). Under the drivers of high demands and high profits, ricecrawfish farming system has recently become increasingly popular among paddy farmers in subtropical China. In the integrated rice-crawfish farming system, resources are well recycled by transferring outputs of a subsystem as inputs of the other system, which generates important co-benefits for nutrient accumulation, productivity, and profitability (Hu et al 2013, Bashir et al 2020, Dong et al 2021, Xu et al 2022). Crawfish in the paddy field provide an immediate source of organic fertilization for crops, and the digging and foraging activities loosen the top soil which facilitates air circulation under the water table (Hou et al 2020, Hu et al 2021). The rice straw and multiple organisms in the farmland also provide shade and food for the crawfish. Most studies used site data from field experiments to investigate the effects on the environment: these predominantly highlighted positive effects on the decrease of greenhouse gas emissions and improvement of nutrients in the soil (Si et al 2017, Xu et al 2021, Gao et al 2022). Our results underscore that the growing importance of rice-crawfish farming calls for accurate spatial and temporal data about the footprint of rice-crawfish farming to characterize its evolution and help understand regional implications. Even though there are rough statistical data about the total area of ricecrawfish cultivation at a regional level, spatial information and change dynamics of this novel agricultural system from emergence to current situation are still lacking (Yu et al 2022). It is therefore urgent to provide distribution data for understanding the spatial knowledge of rice-crawfish farming in China and supporting relative studies on ecological effects. Recent advances in satellite data and computational resources have enabled the mapping of the growing extent of rice-crawfish farming at high spatial and temporal resolutions (Gorelick et al 2017, Weiss et al 2020). Studies on remote-sensed mapping of rice-crawfish system focused on effective identifying methods which are mostly in small regional scales and short time series, nevertheless, spatio-temporal information at a large scale is currently absent which hinders understanding the holistic characteristics of development across China (Wei et al 2019,2022, Xia et al 2021). We here used Landsat imagery to map, for the first time, rice-crawfish production every year from 2013 to 2021 in the middle-lower Yangtze Plain in China, the epicenter of global crawfish production. We then analyzed the characterization of change dynamics and further investigated the effects of ricecrawfish expansion on past land systems. This analysis sheds much-needed light on the development knowledge of a promising and highly efficient integrated crop-aquaculture system that is currently increasing in popularity across China and that may also enrich the product portfolio of small farmers in other regions of the world. 2. Study area We focused on the five provinces in China that produce most crawfish, namely, Anhui, Hubei, Hunan, Jiangsu, and Jiangxi. These provinces are located between 24◦N and 36◦N and 108◦E and 122◦E, and cover 805 600 km2(figure 1). The region has a subtropical climate with hot and humid summers and mild winters and paddy rice cultivation is the dominant farming system. The annual average precipitation is approximately 800–1500 mm; the average temperature ranges between 26 ◦C and 30 ◦C in summer and between −1◦C and 5 ◦C in winter. The Yangtze River, the longest river in Asia, runs through the region. Thousands of smaller lakes are scattered throughout the study area, including the two largest 2 Environ. Res. Lett. 18 (2023) 064040 Y Wei et al Figure 1. Location of the study area. (A): Crawfish production (tons) in China in 2019 at the province level. The study area is the five provinces with the highest production. (B): Elevation gradient and main water resources (in blue) in the study area. freshwater lakes in China, Poyang Lake in Jiangxi and Dongting Lake in Hunan. In 2019, approximately two million tons of crawfish, 92% of China’s domestic production and 88% of global production, were produced in the study area (National Bureau of Statistics of China (NBSC) 2020a). Paddy rice planted in these five provinces occupies 14 million hectares accounting for 48% of China’s total rice cultivation area (National Bureau of Statistics of China (NBSC) 2020b). The river floodplains and lakeshores with widespread paddy fields, sufficient water resource, and suitable climate conditions provide favorable conditions for ricecrawfish cultivation. As the center of crawfish production where rice-crawfish expansion considerably took place, this study area is highly appropriate to understand its development characteristics at a spatial level and comprehensively analyze its change dynamics and effects. Integrated rice-crawfish farming system was developed by local farmers in Hubei Province and the standard technology was initially approved by the China Fisheries Association in 2013 (Cao et al 2017). The production process typically includes two periods per year (figure 2): the rice planting period (from June to October) and the rice fallow period (from November to May of the following year). The field layout of the integrated rice-crawfish cultivation system is characterized by rearing trenches that surround the paddy fields. The trenches are excavated to provide refuge for crawfish when the paddy fields run dry. The crawfish re-enter the fields after the fields are inundated again in June and the young rice stems have grown strong enough to withstand the crawfish. Farmers generally add crawfish two times per year: juvenile crawfish around March, and adult crawfish before the rice harvest. Overall, the typical annual production cycle yields one harvest of rice and at least two harvests of crawfish. 3. Data and methods 3.1. Data sources We used data from the Landsat 8 Operational Land Imager and Landsat 7 enhanced thematic mapper plus (ETM+). Landsat 8 images with high cloud cover were supplemented by Landsat 7 images. We produced composite images from all available imagery for the study area from 2013 to 2021 for the two key farming periods, from July to October and from November to March of the following year. All images have been preprocessed in the Google Earth engine (GEE). We used 1300 points of integrated rice-crawfish farming and 1138 points of other land use types to validate the accuracies of mapping (figure S1). We collected the crawfish sample points through visual interpretation of the available high-resolution images from 2013 to 2021 on Google Earth. The unique spatial pattern of crawfish cultivation, that is, the water trenches surrounding the rice fields, is visible in very-high spatial resolution data (figure S2). When high-resolution imagery was not available from Google Earth for a point, we labeled those points using Landsat surface reflectance data extracted from the GEE Timeseries Explorer plugin in QGIS (Rufin et al 2021). To investigate the effects of expansion for ricecrawfish system on original land-use systems, we use a high-quality land-use dataset in 2010 derived by Resource and Environment Science and Data Center (www.resdc.cn). This dataset is at 30 m spatial resolution and was generated by Landsat TM/ETM+ images, which includes a three-level classification 3 Environ. Res. Lett. 18 (2023) 064040 Y Wei et al Figure 2. The typical temporal cycle of integrated rice-crawfish farming. Cultivation of integrated rice-crawfish system is occupied by two periods: rice cropping and rice fallow. scheme and a sum of 33 classes. We reclassified and focused on five first-level classes and two secondlevel classes, which are irrigated cropland, rainfed cropland, water bodies, woodlands, built-up land, grasslands, and bareland. Detailed description of data sources is provided in the supplementary material. 3.2. Mapping of rice-crawfish cultivation The changes in water bodies between two cultivation periods are essential information for mapping rice-crawfish farming (Wei et al 2021). Based on the object-based water difference method that integrates pixel-based identification with object-based segmentation, we introduced classification and regression tree (CART) decision-tree algorithm and simple non-iterative clustering algorithm (SNIC) in this study (figure S3). We first mapped the distribution of water bodies by using automated water extraction index (AWEIsh) (Feyisa et al 2014). CART decision tree algorithm was used to determine the thresholds of AWEIsh by inputting water samples. We then extracted the integrated rice-crawfish area at the pixel level based on changes in the presence of water between the fallow and rice planting periods. To identify rice-crawfish pixels on land parcels, we used the SNIC algorithm to segment Landsat images for every year on GEE. SNIC clusters are based on the connectivity of pixels and have been tested with fast-running procedures with good performance (Achanta and Susstrunk 2017). We intersected the segmentation results with the rice-crawfish pixels. The percentage of crawfish pixels within each parcel object was calculated and used 50% as the threshold for rice-crawfish detection. We mapped rice-crawfish farming only when there was a change in the presence of water between the fallow and planting periods. By using this method, we generated the spatial distribution of rice-crawfish farming from 2013 to 2021. We assessed the accuracies from 2013 to 2021 by using error matrix and described the overall accuracy, producers’ accuracy and user’s accuracy in each year, and showed good performance with overall accuracies higher than 90%. Details of rice-crawfish mapping method are described in the supplementary material. 3.3. Characterizing of spatio-temporal dynamics Because of the small sizes and scattered distribution of rice-crawfish fields, it was difficult to visualize regional patterns of change dynamics at a 30 m spatial 4 Environ. Res. Lett. 18 (2023) 064040 Y Wei et al resolution in a large study area. We used hexagons instead of squares to highlight the change tendency as visualization with hexagons is more natural and visually appealing besides other benefits (Birch et al 2007, Senf and Seidl 2021). We subdivided the entire study region into a ca. 1000 km2hexagon grid, with a total number of 901 hexagons. Within each hexagon, we counted the number of rice-crawfish pixels per year and calculated the total area of rice-crawfish farming for each hexagon by multiplying by 900 m2. To reveal the characteristics of rice-crawfish distribution from a spatial statistical perspective, we calculated standard deviational ellipse (SDE). SDE is a statistical method to reveal the spatial distribution characteristics of geographic elements from a spatial statistical perspective (Yuill 1971). The azimuth, center coordinates and ellipse area reveal the directionality, center of gravity and concentration and divergence of geographic elements distribution in the research region (Zhang et al 2020). We mapped SDE for the year of 2013, 2015, 2017, 2019, and 2021. The coordinates of the centroids were calculated as follows: X=∑n i=1xi nY=∑n i=1yi n(1) where ¯ X and ¯ Y represent the coordinates of the centers; nis the number of rice-crawfish fields; xiand yi represent the xcoordinate and ycoordinate of each rice-crawfish cultivation field unit, respectively. We quantified the area of rice-crawfish cultivation at a provincial level and calculated the area change rates of the entire region from year to year as follows: Area change rate =(Arean−Arean−1) Arean−1 ×100%(2) where Arean,n−1indicates the area of rice-crawfish farming of the total study region; n represents the year from 2014 to 2021. Areaa,bAiming for quantification of rice-crawfish change dynamics within a fine scale, we calculated the total area amount at a county level and mapped area changes at a two-year interval from 2013 to 2021. Proportions of counties within different ranges of change area were counted to reveal the divergence of change dynamics among years and regions. 3.4. Transformation between rice-crawfish cultivation and other land uses To investigate the effects of rice-crawfish expansion on past land systems, we conducted a transition matrix at a pixel level by using land use and land cover distribution in 2010 and the current rice-crawfish distribution. To analyze the full extent of the impact on rice-crawfish expansion, we chose 2019, the year with the highest area and largest extent of rice-crawfish farming. The classes of land use and land cover systems in 2010 were categorized into seven classes: irrigated cropland, rainfed cropland, water bodies, builtup land, grasslands, woodlands, and bareland. Class of water bodies includes lakes, rivers, and other artificial water areas, such as aquaculture ponds. By calculating the proportion of transformed area amount for different classes at a regional level, we investigated the spatial effects of rice-crawfish farming expansion on land use patterns. To highlight the contribution from main land use types, we selected the top three land use and land cover classes with a high area amount of transformation and subsequently used a hexagon grid to map the spatial characteristics. 4. Results 4.1. Distribution of rice-crawfish farming system In Hubei, Rice-crawfish production has clustered in a few core zones with intensive density in the south of Hubei which is the original region of this system and has experienced a long-term experiment of ricecrawfish farming before 2013 (figure 3(A)). In Hunan and Jiangxi provinces, rice-crawfish cultivation surrounds the largest freshwater lakes in China, the Dongting and Poyang Lakes (figures 4(B) and (C)). The spatial patterns of rice-crawfish cultivation are more scattered in Anhui and Jiangsu and concentrate along the Yangtze River and its branches. Regions with significant increasing trends in rice-crawfish production in 2021 include the west of Anhui and Jiangsu. Rice-crawfish systems are located primarily in low-elevation plain areas with abundant water resources, where paddy rice cultivation has traditionally dominated agriculture (figure 4). At the initial stage when rice-crawfish farming technology was first officially released, rice-crawfish farming was concentrated in the southwestern part of Hubei Province, which is mainly in the Jianghan Plain. With the gradual spread of technology, the influence of ricecrawfish farming become stronger, as shown by the spread of distribution in the north of Hunan and Jiangxi. The extent of rice-crawfish in Anhui and Jiangsu started to enlarge in 2017, however, the area of rice-crawfish in each hexagon is much lower than in other provinces, indicating a scattered distribution and low density. The spatial distribution of SDEs experienced considerable enlargement towards the direction of eastern and northern, especially from 2013 to 2017 (figure 5). Under the large expansion of rice-crawfish cultivation in Jiangsu and Anhui, the distribution of SDE gradually across the boundary of Jiangsu Province in 2019 and 2021. The centers movement route of rice-crawfish was consistently within the boundary of Hubei Province, which has the highest area of rice-crawfish cultivation across the study period. The centers of rice-crawfish distribution moved towards the northeastern direction with 5 Environ. Res. Lett. 18 (2023) 064040 Y Wei et al Figure 3. Annual expansion of integrated rice-crawfish cultivation from 2013 to 2020. Insets (A)–(F) zoom into selected hotspots. Insets (G) and (H) are high-resolution images of rice-crawfish fields in April 2022 and September 2021 (images are screenshots from Google Earth). (G) Shows rice-crawfish farming in the fallow period with inundated fields and (H) with rice planting. Figure 4. Annual expansion of integrated rice-crawfish farming on a hexagonal grid. (A) total of 901 hexagons of approximately 1000 km2each cover the study area. 6 Environ. Res. Lett. 18 (2023) 064040 Y Wei et al Figure 5. Change of standard deviation ellipses and centers movement of rice-crawfish cultivation from 2013 to 2021. Standard deviation ellipses denote the main trend direction and scope of rice-crawfish distribution. The centers movement denotes the change tendency of spatial location for the gravity of distribution. Figure 6. Area of integrated rice-crawfish production in five provinces from 2013 to 2021. Bars show the area at a provincial level, line shows the change rate of rice-crawfish farming area from year to year for the entire region. a total of accumulated 88 km from 2013 to 2021, at an average rate of 11 km per year. 4.2. Spatial-temporal dynamics The total area of rice-crawfish farming in the five provinces increased steadily from 0.11 Mha in 2013 to 0.70 Mha in 2019 (figure 6). Hubei was the largest contributor to rice-crawfish farming in the study region and the area occupied about 60% of the total. Rapid increases in area in five provinces happened since 2017. The area amount reached its peak in 2019, and the largest contribution is from Hubei province (nearly 0.4 Mha), followed by Anhui and Hunan. The change rate consistently increased from 2015 to 2019 and showed a sharp drop in 2020 when a considerable area contraction of 0.22 Mha suddenly happened. The contraction was particularly stark in Hubei Province, with a decrease of 0.09 Mha, followed by Hunan Province and Jiangxi Province. In 2021, the areal extent of rice-crawfish farming partially rebounded to 7 Environ. Res. Lett. 18 (2023) 064040 Y Wei et al Table 1. Distribution of the trends in change area at a county level from 2013 to 2021. Range Proportion of the change area at a county level across five provinces (%) 2013–2015 2015–2017 2017–2019 2019–2021 ⩽−200 0.00 0.00 0.00 0.35 −200 ⩽−60 0.00 0.00 0.00 4.55 −60 ⩽−5 0.00 0.00 0.00 13.99 −5⩽5 78.17 70.69 65.20 61.54 5⩽60 21.40 28.02 28.21 19.23 60 ⩽200 0.44 2.16 6.23 0.35 >200 0.00 0.00 0.37 0.00 Figure 7. Change tendency of rice-crawfish area in five provinces from 2013 to 2021. The figure shows the area change of rice-crawfish cultivation at a county level calculated with a two-year interval. The counties with labels (a), (b), and (c) are Qianjiang, Huoqiu, and Jianli, which are counties with the most notable area change in both positive and negative trends. 0.60 Mha. Anhui province has the largest increase in the rice-crawfish area in 2021 with over 0.04 Mha. Statistically, change area of most counties in five provinces was stable and within an interval between −5 km2and 5 km2, shown by an average occupation of 68.9% (table 1). Besides, increase trends in area from 2013 to 2019 happened within a range of 5– 60 km2, with proportions around 30%. Only several counties had increases over 60 km2, including Qianjiang, Jianli, and Huoqiu (figure 7). From 2017 to 2019, many counties experienced notable increases between 60 km2to 200 km2, as shown by the occupation raised to 6.23%. Different from increase trends dominated until 2019, ca. 20% of counties with negative trends in area amount were found from 2019 to 2021. Most counties with large area decreased from 2019 to 2021 were the same as the counties with a high increase area in 2019, especially Jianli has the strongest negative trend. 4.3. Effects of rice-crawfish expansion on land systems The expansion of rice-crawfish farming underwent a considerable transformation from the original land systems since 2010 (figure 8). Overall, 69% of the ricecrawfish farming fields are transferred from cropland, and the rest are transitioned from other land systems. Within the cropland class, irrigated cropland, which is dominated by paddy rice cultivation, is the largest contributor with a proportion of 56%. Paddy rice fields are generally waterlogged fields with abundant water resources and advanced irrigation facilities systems nearby. Rainfed cropland with no irrigation equipment has a much higher cost to reconstruct 8