Search

Analyzing the temporal hydrological response of regional precipitation to climate variables is critical for improving precipitation prediction and understanding the underlying mechanism. We study the potential relationship between the annual maximum daily precipitation (RX1day) in Guangdong Province and 21 climate variables, including the length of the day of the earth's rotation (LOD), Niño 3.4, and Pacific Decadal Oscillation (PDO). Wavelet coherence (WTC), partial wavelet coherence (PWC), and multiple wavelet coherence (MWC) are used to identify possible individual, independent, and coupled relationships between the RX1day and climate variables. The analysis results show that Tropical Southern Atlantic Index (TSA), Niño 3.4, Southern Oscillation Index (SOI), Western Pacific Index (WP), Tripole Index for the Interdecadal Pacific Oscillation (TPI), North Pacific pattern (NP), Atlantic Multidecadal Oscillation (AMO), and Tropical Northern Atlantic Sea Surface Temperatures Index (TNA) are important driving forces influencing RX1day in Guangdong. After excluding the influence of Niño 3.4 and PDO, PWC accurately explains the influence of a single climate variable on RX1day. A single climate variable is insufficient to describe the influence on RX1day, but MWC accurately reflects the combined influence of different climate variables on RX1day. Our work emphasizes the use of PWC and MWC to reveal the independent and combined influences of climate variables on RX1day on different time scales. This study provides new insights into selecting the optimum predictive factors of RX1day. • We systematically study the possible relationship between 21 climate variables on the time scale. • We discuss the independent influence of climate variables on RX1day after excluding Niño 3.4 and PDO. • We explored the combined influences of different climate variables on RX1day. • Our study has important reference significance for selecting the best predictive factor of extreme precipitation. [ABSTRACT FROM AUTHOR]

Background: Affective states, including sentiment and emotion, are critical determinants of health. However, few studies among men who have sex with men (MSM) have examined sentiment and emotion specifically using real-time social media technologies. Moreover, the explorations on their associations with sexual and health status among MSM are limited.Objective: This study aimed to understand and examine the associations of affective states with sexual behaviors and health status among MSM using public data from the Blued (Blued International Inc) app.Methods: A total of 843,745 public postings of 377,610 MSM users located in Guangdong were saved from the Blued app by automatic screen capture. Positive affect, negative affect, sexual behaviors, and health status were measured using the Simplified Chinese Linguistic Inquiry and Word Count. Emotions, including joy, sadness, anger, fear, and disgust, were measured using the Weibo Basic Mood Lexicon. A positive sentiment score and a positive emotion score were also calculated. Univariate and multivariate linear regression models on the basis of a permutation test were used to assess the associations of affective states with sexual behaviors and health status.Results: A total of 5871 active MSM users and their 477,374 postings were finally selected. Both positive affect and positive emotions (eg, joy) peaked between 7 AM and 9 AM. Negative affect and negative emotions (eg, sadness and disgust) peaked between 2 AM and 4 AM. During that time, 25.1% (97/387) of negative postings were related to health and 13.4% (52/387) of negative postings were related to seeking social support. A multivariate analysis showed that the MSM who were more likely to post sexual behaviors were more likely to express positive affect (beta=0.3107; P<.001) and positive emotions (joy: beta=0.027; P<.001), as well as negative emotions (sadness: beta=0.0443; P<.001 and disgust: beta=0.0256; P<.001). They also had a higher positive sentiment score (beta=0.2947; P<.001) and a higher positive emotion score (beta=0.1612; P<.001). The MSM who were more likely to post their health status were more likely to express negative affect (beta=0.8088; P<.001) and negative emotions, including sadness (beta=0.0705; P<.001), anger (beta=0.0058; P<.001), fear (beta=0.0052; P<.001), and disgust (beta=0.3065; P<.001), and less likely to express positive affect (beta=-0.0224; P=.02). In addition, they had a lower positive sentiment score (beta=-0.8306; P<.001) and a lower positive emotion score (beta=-0.3743; P<.001).Conclusions: The MSM social media community mainly expressed their positive affect in the early morning and negative affect after midnight. Positive affective states were associated with being sexually active, whereas negative affective states were associated with health problems, mostly about mental health. Our finding suggests the potential to deliver different health-related intervention strategies (eg, psychological counseling and safe sex promotion) on a social media app according to the affective states of MSM in real time. [ABSTRACT FROM AUTHOR]