Improved soil fertility to sustain crop productivity is important to enhance agroecosystem services. To address the low nutrient content, subsoil compaction, and poor root penetration in Planosol, a new machine was designed to improve these conditions. This machine integrates subsoil mixing and fertilizer application. The paper details the machine’s structure and working principle, focusing on optimizing the vertical mixing plow. Furthermore, discrete element method simulations were conducted, using tillage depth, forward speed, and plow height as factors, with soil mixing rate and tractive resistance serving as the evaluation indicators. Results showed optimal performance at 553 mm tillage depth, 1 m/s speed, and 227 mm plow height, with tractive resistance of 7460.5 N and a soil mixing rate of 72.1%. Field trials and soil improvement tests based on optimized parameters confirmed the accuracy and reliability of the simulation results. Compared to the Shallow plowing and subsoiling area, the subsoil mixing improvement area showed significant improvement, with a yield increase of 16.4–18.9% over two years.

Albi-boric argosols, mainly distributed in the Sanjiang Plain of Heilongjiang Province, China, accounting for over 80% of the total cultivated land area, is characterized by a nutrient-deficient layer beneath black soil. This study addresses the challenges of modern agriculture by investigating the impact of straw incorporation on soil dissolved organic carbon (DOC) and its structures in albi-boric argosols, profiles, using fluorescence excitation–emission spectroscopy and parallel factor analysis (PARAFAC). Three treatments were applied: undisturbed albi-boric argosols (C), mixed albic and illuvium layers (M), and mixed albic and illuvium layers with straw (MS). Results showed that the yield of M and MS increased by 9.9% and 13.0%, respectively. There was a significant increase in DOC content, particularly in the MS treatment. Fluorescence index (FI) values ranged from 1.65 to 1.86, biological index (BIX) values were less than 1, and humification index (HIX) values were below 0.75, indicating a mix of plant and microbial sources for DOC, autochthonous characteristics, and weaker humification degree. PARAFAC identified two/three individual fluorophore moieties that were attributed to fulvic acid substances, soluble microbial products, and tyrosine-like substances, with microbial products as the dominant component. This study demonstrates the effect of improving barrier soil and maintaining sustainable agriculture by enhancing soil quality.

In this paper, a three-stage subsoil interval mixing four-gang plough (TSIMF) was manufactured to enhance the property of planosol soil in 2016. The working width is set to 2 m to obtain a low operation cost. The total draught of the TSIMF was about 100 kN, the running resistance of the tractor included. The traction force (draught) of TSIMF was determined and the soil condition improved with the TSIMF was discussed. A field test of soybean with TSIMF showed that the soil moisture was about 2% volume greater than that in the CK (subsoiler) field. The soil temperature with the TSIMF was about 1 ºC greater than the CK in the plowed layer, and the yield was 120% of the treatment of the CK. Sanjiang Plain is one of the most important commodity grain bases in China, and the albic soil accounts for about 25% of the total cultivated land. Improving the property of the albic farming land attaches great significance to the stability and increase of national grain production.

Farmland soil water-soluble organic carbon (WSOC), serving as a labile carbon substrate for microbial utilization, demonstrates pronounced sensitivity to land-use modifications and agricultural management practices. This study systematically investigated the impacts of long-term straw incorporation frequencies—including annual (S-1), biennial (S-2), and triennial (S-3) return patterns—on WSOC distribution across 0–20 cm and 20–40 cm soil profiles. Through the integration of three-dimensional excitation–emission matrix (EEM) fluorescence spectroscopy with parallel factor analysis (PARAFAC), we elucidated structural characteristics and humification dynamics associated with different incorporation regimes. The results showed a depth-dependent WSOC distribution pattern with higher concentrations in surface soils (0–20 cm: 261.2–368.9 mg/kg) compared to subsurface layers (20-40 cm: 261.8–294 mg/kg). Straw incorporation significantly increased WSOC content in the 0–20 cm of 16.9%~21.7% and 20–40 cm soil layers of 6.2%~12.3%. Biennial return had the lowest WSOC/SOC ratio, indicating enhanced stability of the soil organic carbon pool. Spectral indices—including the fluorescence index (FI, 1.59~1.69), biological index (BIX, 0.90~0.95), and humification index (HIX, 0.64~0.74)—collectively indicated that WSOC predominantly consisted of microbially processed organic matter with a low degree of humification. PARAFAC modeling resolved two fluorescent components: C1 (humic acid-like substances, 47.4–50.4%), C2 (soluble microbial metabolites, 49.6–52.6%). This systematic investigation provides mechanistic insights into how straw management temporality regulates both quantity and quality of labile carbon pools in agricultural ecosystems.

Based on the analysis of crop growth cycle and water demand, the factors affecting crop growth water use are divided into three categories: environmental factors, crop factors and soil factors. The training set and test set of the model are selected from the crop irrigation historical data set that meets the expected quality and yield. By designing an intelligent farmland irrigation model based on LSTM neural network algorithm, a method of precise irrigation according to crop growth needs, growth environment and planting soil is proposed. According to the characteristics of factors affecting the water consumption for crop growth, the number of hidden layers of the prediction model is determined, and the network parameters are adjusted; The model is trained on the processed historical irrigation data set to obtain the crop irrigation volume prediction model; The LSTM neural network irrigation prediction model is compared with the traditional RNN neural network irrigation prediction model. The experimental results show that the predicted value and trend of LSTM irrigation prediction model are closer to the real value, with stronger robustness, lower error rate and shorter running time, which can meet the prediction of intelligent farmland irrigation and provide reliable basis for the research of intelligent agriculture.