The coastal wetlands of the Yellow River Delta (YRD) in China are crucial for their valuable resources, environmental significance, and economic contributions. However, these also vulnerable to dual threats climate change human disturbances. Despite substantial attention historical shifts YRD's wetlands, uncertainties remain regarding future trajectory face compounded risks from anthropogenic activities. Based on a range remote sensing data sources, this study undertakes comprehensive...

The increasingly frequent and severe regional-scale compound heatwave‒drought extreme events (CHDEs), driven by global warming, present formidable challenges to ecosystems, residential livelihoods, economic conditions. However, uncertainty persists regarding the future trend of CHDEs their insights into regional spatiotemporal heterogeneity. By integrating daily meteorological data from observations in 1961–2022 climate models (GCMs) based on Shared Socioeconomic Pathways, evolution patterns...

Abstract In this paper, reanalysis data collected by ERA‐Interim 4 times daily from 1979 to 2014 are used analyze the main atmospheric moisture sources of cumulative effect torrential rain (CETR) events during preflood season (1 April 30 June) over South China. Using a Lagrangian particle dispersion model—the Hybrid Single‐Particle Integrated Trajectory major 66 CETR for period 1979–2014 identified and compared. study, five sources—the land area, Indian Ocean, Pacific China Sea, China—are...

The long range correlation of extreme events the Lorenz system is discussed by using method fixed threshold. It turns out that all with different threshold have correlation. scaling exponents are similar, but just smaller than original series. less effected initial value changes, it decreases distinctly when parameters increases. systems series has traits memory compared Gaussian white noise. Finally, we use maximal day air temperature data 194 stations between 1957 and 2004, from National...

Radar echo extrapolation based on deep learning is an important method for conducting precipitation nowcasting. sequence data have spatio-temporal correlations and non-rigid movements of the radar echo. According to characteristics data, this study proposes a new fusion neural network called STUNNER. STUNNER implements two-stream strategy extract fuse spatial temporal signals. Specifically, it uses novel cross-network embedding achieve efficient fusion; integrates differencing trajectory...

Based on the idea of using historical-analogue information to revise prediction errors National Climate Centre numerical business model, for North China, based analysis data CMAP from 1983 2009, 40 pieces climate indices NOAA, 27 years season model results 2009 and 74 circulation characteristics materials provided by Weather Diagnostic Forecasting Room Center, method combining simulation diagnostic tests, taking advantage error key similar historical data, identifying factors, optimizing...

In order to enhance the prediction accuracy and computational efficiency of chaotic sequence data, issues such as gradient explosion long computation time traditional methods need be addressed. this paper, an improved Particle Swarm Optimization (PSO) algorithm Long Short-Term Memory (LSTM) neural network are proposed for prediction. The temporal pattern attention mechanism (TPA) is introduced extract weights key information each input feature, ensuring nature historical data. Additionally,...

Abstract The roles of moisture sources and transport in intraseasonal precipitation variabilities during the preflood season over South China are complex different. In this paper, ERA‐Interim reanalysis data that were collected four times daily from 1979 to 2017 used analyse variation influence China. Using Hybrid Single‐Particle Lagrangian Integrated Trajectory model, fixed atmospheric defined: land area source (LD source), Indian Ocean (IO Pacific (PO Sea (SCS (SC source). accounts for...

The circulation of rice from contaminated areas could escalate exposure risk a local problem to national issue and affect wider population beyond the region origin, as confirmed by "Poison Rice Incident" in May 2013 Guangzhou, China. In present study, authors established food chain model based on aquivalence method identify major sources methylmercury (MeHg), estimate levels MeHg, quantify MeHg via aquatic consumption. Different types organism samples Haihe River also were collected verify...

The approximate entropy method is applied to an ideal time series. It indicated that the capable of distinguishing between different structural dynamics and detecting mutation. And then，this observation data daily temperature precipitation 740 sites 1960 2000. results showed that，the distribution has a characteristic obvious district distribution，and displays territorial difference dynamic structures climate system. Abrupt changes have been detected by annual 1970s 1980s. characteristics in...

Based on the theory of probability distribution record-breaking events and daily high/low observational temperature data in China form 1960 to 2005, spatial-temporal characteristics are investigated. We found that annual mean frequency high is about 2.5 times greater than normal 1976—2005, while for low 4.5 less. Record-breaking obviously more Northwest, North China, Northeast Tibet, less there. As relative trend events, somewhat increases most parts but become all over China. For...

Approximate entropy （ApEn） is well known to be an effective abrupt change detection method in dynamic structure. Based on this, we compare the performances between moving detrended fluctuation analysis （MDFA） and ApEn detecting change. The results show that MDFA almost do not depend length of subseries, while could identify structure some extent but still depends subseries. At same time, there exists a huge drift for which means actual time-instants dont coincide with detected one....

This study used long-term in situ rainfall, snow, and streamflow data to explore the predictive contributions of snowmelt rainfall six watersheds over Western United States. Analysis showed that peak snow accumulation, snow-free day, slope all had strong correlation with streamflow, particularly inland basins. Further analysis revealed variation accumulation anomaly lead anomaly. Over entire States, inner mountain areas times 4–10 pentads. However, coastal areas, nearly sites less than one...

This paper takes the typical chaotic system， Lorenz as subject. We use detrended fluctuation analysis method to study systems long-range correlation for different initial values and parameters. It turns out that long range is related with its values phase space position. When value close unstable equilibrium points， rang strengthened， scaling exponents α are bigger， predictability of system better. in complete state， becomes weak parameters increasing， decrease， weaker. reveals...

The occurrence of record breaking temperature events （RBTE） became more and frequent under the background global warming. What is probability that current will be broken in next five years? In this study, answers are provided by given a variety simple statistical models for temperature, relationship between RBTE China warming explored comparative analysis last 50 years. results show frequency influenced autocorrelation. This study quantitatively estimates warming, which can deepen our...

The paper investigates the height and temperature fields reanalysis data of NCEP/NCAR by means detrended fluctuation analysis method, gives their scaling exponent distribution East Asia region. empirical investigation indicates that both field filed have property long range correlation, which matches with features space as a whole. As for grid same layer, correlation varies obviously when latitude changes. low-latitude areas near equator bigger better whereas mid-latitude high smaller worse...