The spatial and temporal variations in global precipitable water (PW) derived from satellite and radiosonde observations during 2007–2016 are compared and analyzed in this study. The two observations show a good agreement in global spatiotemporal distributions and trends in PW. Generally, PW shows apparent latitudinal and seasonal variations. The averaged PW peaks in the tropical region, with the value of 40 mm or higher, and then decreases apparently to the polar region, less than 10 mm. PWs are larger in warm seasons and smaller in cold seasons for both hemispheres. It is indicated that the spatial and seasonal variabilities of PW are highly related to water vapor source and surface temperature as a direct heat source. The global PWs have changed in the past 60 years and, to some extent, strongly in the recent 10 years. It is found that PWs at Australian stations present statistically significant decreasing trends in last 60 years, while PWs at most stations in the United States have increasing trends. According to the global observations, it is found that PWs show uptrends over land and downtrends over the ocean in last 10 years, implying the widespread increase of water vapor in the troposphere over land.

Based on long-term rainfall measurements (1956–2012), water level records (1956–2006), and aquatic plants field survey data (1960–2014), the relationship between precipitation and aquatic vegetation succession in east Taihu Lake, China, is studied. Neither abrupt changes nor any trends were found in the annual rainfall series in Taihu Lake during the studied period (1956–2012). However, for seasonal variations, statistically significant decreases are found in spring and autumn, while the rainfall in winter exhibits statistically significant increase. No significant trend was obtained in summer. A “dry” period was detected in our studied period (1963/1964~1978/1979). Total annual rainfall was significantly positively correlated to the number of rain-days (r=0.59) and the water level (r=0.84). Our results indicate that the variations of rainfall and water level may have an impact on the aquatic plants in Taihu Lake. The dry period may be not suitable for the growth of the aquatic plants. All aquatic plants in Taihu Lake were dramatically reduced in the dry period, especially for submerged macrophytes and floating-leaf macrophytes. Our results may be helpful for the aquatic restoration in the future.

In this study, Precipitation Estimation from Remotely Sensed Information Using Artificial Neural Networks–Climate Data Record (PERSIANN-CDR) was evaluated for the assessment of long-term drought monitoring in Huaihe River Basin using daily gauge observation data for the period from 1983 to 2017. The evaluation results show that the PERSIANN-CDR algorithm has a good detection ability for small precipitation events over the whole basin, but a poor ability for extreme precipitation events (>50 mm/day). Daily PERSIANN-CDR estimates perform relatively better in areas with abundant precipitation, while the monthly and yearly PERSIANN-CDR estimates are highly consistent with gauge observations both in magnitude and space. The Standardized Precipitation Index (SPI) at various time scales (3, 6, and 12 months) was calculated based on PERSIANN-CDR and gauge observation, respectively. Grid-based values of statistics derived from those SPI values demonstrate that PERSIANN-CDR has a good ability to capture drought events of each time scale across the basin. However, caution should be applied when using PERSIANN-CDR estimates for basin-scale drought trend analysis. Furthermore, three drought events with long duration and large extent were selected to test the applicability of PERSIANN-CDR in drought monitoring. The results show that it has a good ability to capture when and where droughts occur and how far they spread. Due to the overestimation of small precipitation events, PERSIANN-CDR tends to overestimate the number of extreme droughts and their extents. This needs to be considered in future algorithm improvement.

With the intensification of human activities over the past three decades in China, adverse effects on river ecosystem become more serious especially in the Hanjiang River. Xiangyang site is an important spawn ground for four domestic fishes in the downstream region of Hanjiang River. Based on the field survey results of macrophytes during 1997–2000 and 2013-2014, community succession of aquatic macrophytes at Xiangyang site was evaluated and discussed. Two-key ecologic-related hydrologic characteristics, flow regime and water level, were identified as the main influence factors. The EFC (environmental flow components) parameters were adopted to evaluate the alteration of flow regimes at Xiangyang site during 1941–2013. Evaluation results demonstrate a highly altered flow process after being regulated by reservoir. The flow patterns tend to be an attenuation process with no large floods occurring but a higher monthly low flow. Furthermore, the water level decreased and fluctuation reduced after the dam was built, which caused the decrease of biomass but favored the submerged macrophytes during 1995–2009. However, with the water level increasing after 2010 and gently fluctuating, due to uplift by the hydraulic projects downstream as well as the flow attenuation, the dominant position of submerged macrophytes will be weakened.

According to the characteristics of optimal operation of cascade hydropower stations, an improved method is presented to solve the multi-stage optimum mathematics model, namely the coarse grain parallel particle swarm optimization (PPSO) algorithm. In order to test its performance, two algorithms are applied to optimal operation of cascade reservoirs in Yalong River, including particle swarm optimization (PSO). The comparison indicated that the results obtained by the PPSO have faster calculation speed and higher accuracy of solution than PSO, moreover, the PPSO can better improve the convergence ability of PSO.