Quantifying crop residue cover (CRC) on field surfaces is important for monitoring the tillage intensity and promoting sustainable management. Remote-sensing-based techniques have proven practical for determining CRC, however, the methods used are primarily limited to empirical regression based on crop residue indices (CRIs). This study provides a systematic evaluation of empirical regressions and machine learning (ML) algorithms based on their ability to estimate CRC using Sentinel-2 Multispectral Instrument (MSI) data. Unmanned aerial vehicle orthomosaics were used to extracted ground CRC for training Sentinel-2 data-based CRC models. For empirical regression, nine MSI bands, 10 published CRIs, three proposed CRIs, and four mean textural features were evaluated using univariate linear regression. The best performance was obtained by a three-band index calculated using (B2 − B4)/(B2 − B12), with an R²_cv of 0.63 and RMSE_cv of 6.509%, using a 10-fold cross-validation. The methodologies of partial least squares regression (PLSR), artificial neural network (ANN), Gaussian process regression (GPR), support vector regression (SVR), and random forest (RF) were compared with four groups of predictors, including nine MSI bands, 13 CRIs, a combination of MSI bands and mean textural features, and a combination of CRIs and textural features. In general, ML approaches achieved high accuracy. A PLSR model with 13 CRIs and textural features resulted in an accuracy of R²_cv = 0.66 and RMSE_cv = 6.427%. An RF model with predictors of MSI bands and textural features estimated CRC with an R²_cv = 0.61 and RMSE_cv = 6.415%. The estimation was improved by an SVR model with the same input predictors (R²_cv = 0.67, RMSE_cv = 6.343%), followed by a GPR model based on CRIs and textural features. The performance of GPR models was further improved by optimal input variables. A GPR model with six input variables, three MSI bands and three textural features, performed the best, with R²_cv = 0.69 and RMSE_cv = 6.149%. This study provides a reference for estimating CRC from Sentinel-2 imagery using ML approaches. The GPR approach is recommended. A combination of spectral information and textural features leads to an improvement in the retrieval of CRC.

Constructed wetlands can play an active role in improving the water quality of urban rivers. In this study, a sequential series system of the floating-bed constructed wetland (FBCW), horizontal subsurface flow constructed wetland (HSFCW), and surface flow constructed wetland (SFCW) were constructed for the urban river treatment in the cold regions of North China, which gave full play to the combined advantages. In the Yitong River, the designed capacity and the hydraulic loading of the system was 100 m³/d and 0.10 m³/m²d, respectively. The hydraulic retention time was approximately 72 h. The monitoring results, from April to October in 2016, showed the multiple wetland ecosystem could effectively remove chemical oxygen demand (COD), ammonia nitrogen (NH₄⁺-N), total nitrogen (TN), total phosphate (TP), and suspended solids (SS) at average removal rates of 74.79%, 80.90%, 71.12%, 78.44%, and 91.90%, respectively. The removal rate of SS in floating-bed wetland was the largest among all the indicators (80.24%), which could prevent the block of sub-surface flow wetland effectively. The sub-surface flow wetland could remove the NH₄-N, TN, and TP effectively, and the contribution rates were 79.20%, 64.64%, and 81.71%, respectively. The surface flow wetland could further purify the TN and the removal rate of TN could reach 23%. The total investment of this ecological engineering was $12,000. The construction cost and the operation cost were $120 and $0.02 per ton of polluted water, which was about 1/3 to 1/5 and 1/6 to 1/3 of the conventional sewage treatment, respectively. The results of this study provide a technical demonstration of the restoration of polluted water in urban rivers in northern China.

Peatlands are organic-matter-rich but nitrogen-limited natural systems, the carbon/nitrogen (C/N) status of which are subject to increasing exposure from long-term nitrate (NO₃⁻) fertilizer inputs and atmospheric nitrogen (N) deposits. To manage and protect these unique environments, an improved understanding of denitrification-dependent anaerobic oxidation of methane (DAMO) in peatlands is needed. In this study, we used stable isotope measurements and incubation with NO₃⁻additions to facilitate an investigation and comparison of the potential DAMO rates in a paddy-peatland that has been influenced by N fertilizer over 40 years and an undisturbed peatland in northeast China. Monitoring of¹³CO₂production confimed DAMO did occur in both the paddy-peatland and the undisturbed peatland, the rates of which increased with NO₃⁻additions, but decreased logarithmically with time. When NO₃⁻was added, there were no significant differences between the CH₄oxidation in the paddy-peatland and peatland samples after 36 hours of incubation (97.08vs. 143.69 nmol g⁻¹dry peat) and the potential DAMO rate after incubation for 1 hour (92.53vs. 69.99 nmol g⁻¹h⁻¹). These results indicate that the occurrence of DAMO in peatlands might be controlled by the amount of NO₃⁻applied and the depth to which it penetrates into the anoxic layer.

Understanding the characteristics of the balanced nutrient requirements for peanut to achieve target yields is paramount when formulating fertilizer management strategies to increase yields and avoid fertilizer loss. Nutritional requirement estimation models can provide effective alternatives for the estimation of the optimum crop balanced nutrient requirements under varied agricultural conditions which are less time consuming and expensive. In the present study, the quantitative estimation of the optimum crop balanced nutrient requirements of peanut in China were obtained using quantitative evaluation of fertility of tropical soils (QUEFTS) model. The database covered the main agro-ecological region for peanut crops in China between 1993 and 2018. The predicted results of the QUEFTS model indicated that nutrient uptake requirements increased linearly with increasing pod yields until the yields had reached approximately 60% to 70% of the potential pod yields. It was found that with the increasing pod yields during the nutrient linear absorption stage, the plants had required 38.4 kg N, 4.3 kg P, and 14.0 kg K in total to produce 1000 kg of pods, and the corresponding internal efficiencies were 26.0 kg N/kg, 235.0 kg P/kg, and 71.6 kg K/kg, respectively. In addition, the balance rates of the removal nutrient in the pods were determined to be 29.4 kg N, 2.9 kg P, and 4.9 kg K per 1000 kg of pod yield, or approximately 76.5%, 67.4%, and 34.7% of N, P, and K in the total plants, respectively. This study’s field validation experiments verified the applicability and accuracy of the QUEFTS model. Therefore, it was considered to be an effective alternative for the estimation of the optimal balance N, P, and K uptake requirements for peanut crops. These findings will potentially be helpful when making future decisions regarding fertilizer recommendations for peanut crops in China.