Search

In high-throughput multi-beam satellite systems, the dimensionality of the spectrum resource allocation problem increased drastically with the number of satellite beams and service users, which caused an exponential rise in the complexity of the solution. To address the challenge, a two-stage algorithm that combined behavior cloning (BC) with deep reinforcement learning (DRL) was proposed. In the first stage, the strategy network was pretrained using existing decision data from satellite operation through behavior cloning, which mimicked expert behavior to reduce blind exploration and accelerate algorithm convergence. In the second stage, the strategy network was further optimized using the proximal policy optimization (PPO), and a convolutional block attention module (CBAM) was employed to better extract the user traffic features, thereby enhancing overall algorithm performance. Simulation results demonstrate that the proposed algorithm outperforms the benchmark algorithms in terms of convergence speed and algorithm stability, and also delivers superior performance in system delay, average system satisfaction, and spectrum efficiency.

Straw returning may stimulate soil microbial activity, thereby influencing microbial-mediated soil nitrification, which can lead to nitrate leaching and nitrogen (N) loss. However, its effects under long-term nitrogen fertilization remain unclear. At an experimental station with 34 years of fertilizer application (0, 135, and 270 kg ha−1 N), we investigated how nitrogen fertilization and straw returning affected the soil potential nitrification rate (PNR) and ammonia-oxidizing microorganisms (AOM). Our results suggest that N fertilization concurrently inhibits soil PNR, but this inhibition can be alleviated by straw returning, particularly with low nitrogen fertilization (p < 0.05). Long-term N fertilization significantly decreased the abundance of ammonia-oxidizing archaea (AOA), ammonia-oxidizing bacteria (AOB), and complete ammonia-oxidizing bacteria cladeB (CAOB-cladeB). Straw returning increased AOA abundance and diversity, especially with low or no fertilization (p < 0.05). Furthermore, the partial least squares path model demonstrated that AOA abundance affected soil PNR by altering the AOA community. According to random forest analysis, soil pH and AOA beta diversity were the primary factors affecting soil PNR (explaining 10.76% and 10.03% of the variation, respectively). Overall, our findings highlight the importance of straw returning and AOA in soil nitrification under long-term nitrogen fertilization, emphasizing the need to consider these interactions for sustainable agriculture. [ABSTRACT FROM AUTHOR]

Microbial necromass carbon (MNC) is a major contributor to soil organic carbon (SOC) sequestration. Fertilization combined with plastic film mulching, as an intensive agricultural practice to increase crop yields, affects soil microbial growth and metabolism. Nevertheless, how fertilization combined with mulching affects SOC sequestration by mediating MNC dynamics remains elusive. Here, the mulching and no‐mulching sub‐treatments were set under three fertilization treatments (no fertilization, NF; inorganic fertilization, IF; manure, MF), and a 900‐day field incubation experiment using polyvinyl chloride containers was conducted in the buffer zone of NF treatment. We investigated the effects of fertilization combined with mulching on MNC composition (including fungal necromass carbon, FNC; and bacterial necromass carbon, BNC) and qualified their contributions to SOC sequestration. The MF treatment with/without mulching significantly increased the contents of MNC, FNC, BNC, and SOC by 97%–122%, 81%–94%, 152%–210%, and 60%–70% compared with the NF treatment without mulching over 900 days, respectively. The MNC content had a strongly positive correlation with particulate organic carbon and microbial biomass C (p < 0.01). During the incubation stage, the proportion of MNC in SOC was higher in the IF (37%–42%) and MF (40%–44%) treatments with/without mulching than that in the NF treatment with/without mulching (31%–35%). On day 900, mulching significantly decreased the MNC content by 8.7% and 7.8%, and decreased the proportion of MNC in SOC by 5.1% and 5.8% under the NF and IF treatments, respectively. In contrast, mulching did not significantly affect the MNC and SOC contents, but significantly increased the FNC content by 4.8% under the MF treatment on day 900. Mulching significantly decreased the proportion of BNC in SOC regardless of fertilization, and increased the proportion of FNC in SOC under the MF treatment on day 900. Overall, our findings suggest that mulching under manure application maintains SOC sequestration by promoting FNC retention. [ABSTRACT FROM AUTHOR]