In recent years, the increased intensity and duration of droughts have dramatically altered the structure and function of grassland ecosystems, which have been forced to adapt to this change in climate. Combinations of global change drivers such as elevated atmospheric CO₂ concentration, warming, nitrogen (N) deposition, grazing, and land-use change have influenced the impact that droughts have on grassland C cycling. This influence, to some extent, can modify the relationship between droughts and grassland carbon (C) cycling in the multi-factor world. Unfortunately, prior reviews have been primarily anecdotal from the 1930s to the 2010s. We investigated the current state of the study on the interactive impacts of multiple factors under drought scenarios in grassland C cycling and provided scientific advice for dealing with droughts and managing grassland C cycling in a multi-factor world. Currently, adequate information is not available on the interaction between droughts and global change drivers, which would advance our understanding of grassland C cycling responses. It was determined that future experiments and models should specifically test how droughts regulate grassland C cycling under global changes. Previous multi-factor experiments of current and future global change conditions have studied various drought scenarios poorly, including changes in precipitation frequency and amplitude, timing, and interactions with other global change drivers. Multi-factor experiments have contributed to quantifying these potential changes and have provided important information on how water affects ecosystem processes under global change. There is an urgent need to establish a systematic framework that can assess ecosystem dynamic responses to droughts under current and future global change and human activity, with a focus on the combined effects of droughts, global change drivers, and the corresponding hierarchical responses of an ecosystem.

Neural networks, especially the latest deep learning, have exhibited good ability in estimating surface parameters from satellite remote sensing. However, thorough examinations of neural networks in the estimation of land surface temperature (LST) from satellite passive microwave (MW) observations are still lacking. Here, we examined the performances of the traditional neural network (NN), deep belief network (DBN), and convolutional neural network (CNN) in estimating LST from the AMSR-E and AMSR2 data over the Chinese landmass. The examinations were based on the same training set, validation set, and test set extracted from 2003, 2004, and 2009, respectively, for AMSR-E with a spatial resolution of 0.25°. For AMSR2, the three sets were extracted from 2013, 2014, and 2016 with a spatial resolution of 0.1°, respectively. MODIS LST played the role of “ground truth” in the training, validation, and testing. The examination results show that CNN is better than NN and DBN by 0.1–0.4 K. Different combinations of input parameters were examined to get the best combinations for the daytime and nighttime conditions. The best combinations are the brightness temperatures (BTs), NDVI, air temperature, and day of the year (DOY) for the daytime and BTs and air temperature for the nighttime. By adding three and one easily obtained parameters on the basis of BTs, the accuracies of LST estimates can be improved by 0.8 K and 0.3 K for the daytime and nighttime conditions, respectively. Compared with the MODIS LST, the CNN LST estimates yielded root-mean-square differences (RMSDs) of 2.19–3.58 K for the daytime and 1.43–2.14 K for the nighttime for diverse land cover types for AMSR-E. Validation against the in-situ LSTs showed that the CNN LSTs yielded root-mean-square errors of 2.10–4.72 K for forest and cropland sites. Further intercomparison indicated that ~50% of the CNN LSTs were closer to the MODIS LSTs than ESA’s GlobTemperature AMSR-E LSTs, and the average RMSDs of the CNN LSTs were less than 3 K over dense vegetation compared to NASA’s global land parameter data record air temperatures. This study helps better the understanding of the use of neural networks for estimating LST from satellite MW observations.

Flood disasters are one of the most serious meteorological disasters in China. With the rapid development of information technology, individual monitoring tools could not meet the need for flood disaster monitoring. Therefore, a new integrated air-space-ground method, based on combined satellite remote sensing, unmanned aerial vehicle remote sensing and field measurement technology, has been proposed to monitor and assess flood disasters caused by a dam failure in Poyang County, Jiangxi Province. In this paper, based on an air-space-ground investigation system, the general flooded areas, severely affected areas, and more severely affected areas were 53.18 km², 12.61 km² and 6.98 km², respectively. The size of the dam break gap was about 65 m and 34.7 m on 22 and 23 June. The assessment precision was better than 98%, and the root mean square error (RMSE) was 0.86 m. The method could meet the needs for flood disaster information at different spatiotemporal scales, such as macro scale, medium scale and local small scale. The integrated monitoring of flood disasters was carried out to provide the whole process and all-round information on flood evolution dynamics, the disaster development process for flood disaster monitoring and emergency assessment, and holographic information for emergency rescue and disaster reduction, as well as to meet the need for different temporal and spatial scales of information in the process of disaster emergencies.

Urban rainstorm waterlogging is one of the most important problems in urban development and a comprehensive embodiment of urban diseases. China is facing a severe risk of rainstorm waterlogging disasters, which is affecting sustainable development. Urban rainstorm waterlogging in China is caused by many factors, including natural factors and human factors, such as climate warming, unreasonable urban construction, inadequate upgrading of urban fortification standards, etc. Based on the analysis of the current strategies to deal with urban waterlogging around the world, including an increase in surface infiltration, and a reduction in runoff (and its various impacts), this paper holds that the connotation and goal of these measures are highly consistent with the construction of a sponge city in China. Based on the analysis of the problems, including construction of an urban rainwater recovery system, construction of urban rainwater storage facilities, and construction of data platforms faced by China’s sponge city, this paper puts forward the guiding principles of promoting the construction of a sponge city. The guiding principles are to cooperate to deal with climate change and ecological civilization construction, to study the foreign experience, and to unite multiple subjects, integrate multiple elements, design multiple processes, form a joint force, and create an all-round response system to deal with urban rainstorm waterlogging. Then, this paper gives policy recommendations on how to deal with the urban rainstorm waterlogging disasters, which include improving the defense standards, encouraging social participation, popularizing the construction of sponge cities, perfecting the monitoring and early warning system, strengthening the scientific planning of cities, strengthening the ability of dealing with catastrophes in metropolitan areas, the overall planning of cross-regional responses, and enhancing the awareness of decision makers. Finally, this paper expounds the reference significance of urban rainstorm waterlogging control in China to the global audience. This paper explores the significance of comprehensively and scientifically understanding urban rainstorm waterlogging disasters, and provides support for long-term planning and high-quality construction of future safe cities.