Experimental design

The experiment began in June 2010. Two planting systems, a traditional planting platform (TPP) as the control and a VEE-IPBP system, were established⁴⁷. For the VEE-IPBP system, both dry and wet fields were coupled within the same space and time (Fig. ^S1). In the dry field, dry-farmed vegetables (taro intercropped with broccoli, The collection of plant materials conforms to local and national standards and regulations, such as Code of practice for Production techniques of taro (DB42/T 1028-2014), Production technical practice for broccoli(GB/Z 26586-2011)) were planted, and earthworms (Pheretima guillelmi, native to Chongming Island) were reared. In the wet field, Monopterus albus (swamp eels) were reared. In this system, residue including the leaves and stalks of the vegetables were returned to and decomposed in the field as food for the earthworms. The earthworms were used as feed for M. albus. The faeces of the earthworms and M. albus provided nutrients for the vegetables. Hence, internal circulation of resources within the agricultural system was achieved⁴⁸. There were 3 replicate plots for each treatment, and the plots were arranged in a randomized block. A commercial organic fertilizer consisting of organic matter (413.4 g/kg), nitrogen (N) (17.1 g/kg), phosphorous pentoxide (12.4 g/kg) and potassium oxide (12.3 g/kg) was applied as basal fertilizer at a dose of 18 t/hm², and a compound fertilizer (N:P₂O₅:K₂O = 15:15:15) (90% as base fertilizer and 10% as a topdressing material) was evenly sprayed at a dose of 375.0 kg/ha² onto the vegetable field surface. The same commercial organic fertilizer was applied as basal fertilizer at a dose of 15 t/ha², and compound fertilizer (N:P₂O₅:K₂O = 15:15:15) (60% as base fertilizer and 40% as topdressing) was evenly sprayed at a dose of 750 kg/ha² onto the cauliflower field surface. Earthworms (each weighing 3 g) were introduced at a density of 120 per m² (the natural density was 60–80 per m² in the surrounding vegetable field). The earthworm species used was William’s worm, the native species of Chongming Island in Shanghai. Over the course of the experiment, new earthworms were born, and old earthworms were killed or eaten by eels. In late winter and early spring when the temperature was above 6–10 °C, the number of earthworms in the field was investigated, and earthworms were removed or added so that the density of earthworms in the soil under the combination of planting and aquaculture was kept at approximately 120/m². The results of the investigation in 2016 showed that the density of earthworms was 155/m² in cultivated plots and 66/m² in noncultivated plots. The yields and economic benefits from the two cropping patterns of earthworm, eel, taro and cauliflower are shown in Table ^S2. Compared with the yield, the economic benefit of the combined model was higher than that of the traditional model.