Effect of partial replacement of inorganic N with organic manure on crop yield and soil nutrient balance in arable ecosystem in karst peak-cluster depression

Based on a long-term experiment on maize-soybean relay intercropping system in the karst peak-cluster depression in Guangxi, the effects of organic nitrogen (manure) in place of chemical nitrogen (fertilizer) on crop yield and soil nutrient balance were investigated. The experiment was started in 2006, and the data were collected in 2010-2014. Four treatments with four repetitions were designed in the study—including CK (no fertilizer), NPK (chemical fertilizer in maize at N 200 kg·hm^-2, P₂O₅ 90 kg·hm^-2, K₂O 120 kg·hm^-2; and in soybean at N 22.5 kg·hm^-2, P₂O₅ 60 kg·hm^-2, K₂O 67.5 kg·hm^-2), C7M3 (the total amounts of N, P and K were same as NPK treatment, in which 70% N was from chemical fertilizer and 30% N from organic cattle manure), and then C4M6 (60% N was from organic cattle manure and 40% N from chemical fertilizer, the amounts of P and K were the same as the treatment NPK). The results showed that: 1) compared with CK, treatments of NPK, C7M3 and C4M6 all increased soil organic matter (SOM), total nitrogen (TN), available phosphorus (AP) and available potassium (AK). Furthermore, SOM content in C4M6 was significantly higher than that in NPK (P < 0.05). The contents of TN, AP and AK increased with increasing organic manure supplement. 2) The yields of maize and soybean in the fertilizer treatments were respectively 4.15-4.36 and 2.47-2.58 times higher than that in CK. Crop yield order in fertilizer treatments was C4M6 > NPK > C7M3, but there was no significant difference among treatments (P > 0.05). 3) For CK treatment, maize yield decreased by 5.45 g·m^-2·a^-1, while soybean yield increased by 1.50 g·m^-2·a^-1 during experimental period. The yields of both maize and soybean increased under long-term fertilization as the experiment went on. 4) In the experiment, K deficit was observed in organic manure treatments and CK during maize growth period, but N deficit appeared in all treatments during soybean growth period. In the maize-soybean relay intercropping system, K deficit of 7.9 kg·hm^-2·a^-1 was observed only in C4M6. P surplus was observed for all fertilizer treatments with surplus amounts of 81.2 kg(P)·hm^-2, 83.4 kg(P)·hm^-2 and 74.8 kg(P)·hm^-2 in NPK, C7M3 and C4M6 treatments, respectively. In summary, based on the characteristics of nutrient balance and crop yield, that partial replacement of chemical nitrogen fertilizer by cattle manure was reasonable. We recommend that farmers should "reduce N, maintain P and increase K" during maize growth period, but should "maintain N and reduce P and K" in soybean growth period in maize-soybean agricultural ecosystem in karst peak-cluster depression areas.