The aim of this study was to extract medium-term nitrous oxide (N₂O) fluctuations, that is fluctuations over time scales of several days, from continuous data in a biological reactor. Although N₂O emission may fluctuate over the medium-term, a large short-term fluctuation made it difficult to observe. We analyzed approximately one month of continuous dissolved N₂O data in a Carrousel reactor for sewage water in 2019. After confirming the characteristics of the data, we applied a seasonal-trend decomposition procedure with loess: STL decomposition, which is one of the nonparametric seasonal adjustment methods. As a result, medium-term fluctuations of dissolved N₂O concentrations were observed as time-series data eliminating short-term diurnal fluctuations. Furthermore, a characteristic peak of the medium-term fluctuation was explained by the fluctuation of the influent volume depending on the immediately preceding rainfall. Therefore, it is necessary to consider the effect of rainfall for a statistical model of nitrous oxide production.

Although there are examples of research aiming to automate the microscopy inspection conducted in the maintenance of activated sludge treatment, the challenge lies in the necessity to collect a large number of images of various microfauna species as training data for model construction. In this study, we aimed to develop an image analysis model that eliminates the need for the annotation of microfauna species and conducts comprehensive detection and classification of such species based on their appearances. Specifically, we constructed a deep learning model composed of a microfauna detector, a feature extractor, and a microfauna classifier, utilizing knowledge of object detection and self-supervised learning. We then quantitatively evaluated the accuracies of the detection and classification of microfauna species and the relationship between the microfauna community composition obtained using this model and the bacterial community composition in activated sludge flocs determined by 16S rDNA amplicon analysis. The microfauna detector achieved AP₅₀ = 82.85% performance, and the microfauna classifier, using supervised learning, achieved 83.3% accuracy, excluding half of the 12 false positive images containing non-microfauna objects. Furthermore, we found a significant correlation between the microfauna community composition obtained using the model applied to 12 activated sludge samples and the bacterial community composition obtained by the rDNA analysis.

At present, many efforts are being made to ensure the biodiversity and productivity of the marine ecosystems in Harima Nada, the Seto Inland Sea, by increasing the land-derived load of nutrients, such as total nitrogen (TN) into the sea. In this study, a food chain model was used to predict the effects of changes in the input of land-derived TN on the marine TN concentration and the Harima Nada ecosystems’ production of phytoplankton, zooplankton, and higher-trophic-level organisms. The model’s ecosystem parameters were estimated stochastically by the Monte Carlo method. The model successfully reproduced the TN concentration and the biomass production in both northern-coastal and central-southern area with different environmental characteristics. By incrementing land-derived TN load 10 times of the 2010s by the model, the estimation of TN concentration in marine waters increases by up to 1.9 times in the northern-coastal area and 1.3 times in the central-southern area. Additionally, there is an increase in the biomass of piscivorous species (highest ecosystem production level) of 1.2 times for the entire Harima Nada.