The large concentration of the world’s population in cities, along with rapid urbanization, have brought numerous environmental and socioeconomic challenges to sustainable urban systems (SUS). However, current SUS studies focus heavily on ecological aspects, rely on SUS indicators that are not supported by available data, lack comprehensive analytical frameworks, and neglect SUS regional differences. This paper develops a novel approach to assessing urban sustainability from regional perspectives using commonly enumerated socioeconomic statistics. It integrates land use and land cover change data and ecosystem service values, applies data mining analytics to derive SUS indicators, and evaluates SUS states as trade-offs among relevant SUS indicators. This synthetic approach is called the integrated socioeconomic and land-use data mining–based multi-objective assessment (ISL-DM-MOA). The paper presents a case study of urban sustainability development in cities and counties in Inner Mongolia, China, which face many environmental and sustainable development problems. The case study identifies two SUS types: (1) several large cities that boast well-developed economies, diversified industrial sectors, vital transportation locations, good living conditions, and cleaner environments; and (2) a few small counties that have a small population, small urban construction areas, extensive natural grasslands, and primary grazing economies. The ISL-DM-MOA framework innovatively synthesizes currently available socioeconomic statistics and environmental data as a unified dataset to assess urban sustainability as a total socio-environmental system. ISL-DM-MOA deviates from the current indicator approach and advocates the notion of a data-mining-driven approach to derive urban sustainability dimensions. Furthermore, ISL-DM-MOA diverges from the concept of a composite score for determining urban sustainability. Instead, it promotes the concept of Pareto Front as a choice set of sustainability candidates, because sustainability varies among nations, regions, and locations and differs between political, economic, environmental, and cultural systems.

Whole-plant corn silage is a predominant forage for livestock that is processed in Heilongjiang province (Daqing city and Longjiang county), Inner Mongolia Autonomous Region (Helin county and Tumet Left Banner) and Shanxi province (Taigu and Shanyin counties) of North China; it was sampled at 0, 5, 14, 45 and 90 days after ensiling. Bacterial community and fermentation quality were analysed. During fermentation, the pH was reduced to below 4.0, lactic acid increased to above 73 g/kg DM (p < 0.05) and Lactobacillus dominated the bacterial community and had a reducing abundance after 14 days. In the final silages, butyric acid was not detected, and the contents of acetic acid and ammonia nitrogen were below 35 g/kg DM and 100 g/kg total nitrogen, respectively. Compared with silages from Heilongjiang and Inner Mongolia, silages from Shanxi contained less Lactobacillus and more Leuconostoc (p < 0.05), and had a separating bacterial community from 14 to 90 days. Lactobacillus was negatively correlated with pH in all the silages (p < 0.05), and positively correlated with lactic and acetic acid in silages from Heilongjiang and Inner Mongolia (p < 0.05). The results show that the final silages had satisfactory fermentation quality. During the ensilage process, silages from Heilongjiang and Inner Mongolia had similar bacterial-succession patterns; the activity of Lactobacillus formed and maintained good fermentation quality in whole-plant corn silage.

With advances in remote sensing, massive amounts of remotely sensed data can be harnessed to support land use/land cover (LULC) change studies over larger scales and longer terms. However, a big challenge is missing data as a result of poor weather conditions and possible sensor malfunctions during image data collection. In this study, cloud-based and open source distributed frameworks that used Apache Spark and Apache Giraph were used to build an integrated infrastructure to fill data gaps within a large-area LULC dataset. Data mining techniques (k-medoids clustering and quadratic discriminant analysis) were applied to facilitate sub-space analyses. Ancillary environmental and socioeconomic conditions were integrated to support localized model training. Multi-temporal transition probability matrices were deployed in a graph-based Markov–cellular automata simulator to fill in missing data. A comprehensive dataset for Inner Mongolia, China, from 2000 to 2016 was used to assess the feasibility, accuracy, and performance of this gap-filling approach. The result is a cloud-based distributed Markov–cellular automata framework that exploits the scalability and high performance of cloud computing while also achieving high accuracy when filling data gaps common in longer-term LULC studies.