Monthly Rainfall Signatures of the North Atlantic Oscillation and East Atlantic Pattern in Great Britain

Atmospheric-oceanic circulations (teleconnections) have an important influence on regional climate. In Great Britain, the North Atlantic Oscillation (NAO) has long been understood as the leading mode of climate variability, and its phase and magnitude have been found to influence regional rainfall in previous research. The East Atlantic Pattern (EA) is also increasingly recognised as being a secondary influence on European climate. In this study we use high resolution gridded rainfall and Standardised Precipitation Index (SPI) time series data for Great Britain to map the monthly rainfall signatures of the NAO and EA over the period January 1950–December 2015. Our analyses show that the influence of the two teleconnections varies in space and time with distinctive monthly signatures observed in both average rainfall/SPI-1 values and incidences of wet/dry extremes. In the winter months the NAO has a strong influence on rainfall and extremes in the north-western regions. Meanwhile, in the southern and central regions stronger EA-rainfall relationships are present. In the summer months opposing positive/negative phases of the NAO and EA result in stronger wet/dry signatures which are more spatially consistent. Our findings suggest that both the NAO and EA have a prominent influence on regional rainfall distribution and volume in Great Britain, which in turn has implications for the use of teleconnection forecasts in water management Decemberision making. We conclude that accounting for both NAO and EA influences will lead to an enhanced understanding of both historic and future spatial distribution of monthly precipitation.

Extreme rainfall deficits were not the cause of recurring colonial era famines of southern Indian semi-arid regions

Using information contained in the eighteenth to twentieth century British administrative documents, preserved in the National Archives of India (NAI), we present a 218-year (1729–1947 AD) record of socioeconomic disruptions and human impacts (famines) associated with ‘rain failures’ that affected the semi-arid regions (SARs) of southern India. By mapping the southern Indian famine record onto long-term spatiotemporal measures of regional rainfall variability, we demonstrate that the SARs of southern India repeatedly experienced famines when annual rainfall reduced by ~ one standard deviation (1 SD), or more, from long-term averages. In other words, ‘rain failures’ listed in the colonial documents as causes of extreme socioeconomic disruptions, food shortages and human distress (famines) in the southern Indian SARs were fluctuations in precipitation well within the normal range of regional rainfall variability and not extreme rainfall deficits (≥ 3 SD). Our study demonstrates that extreme climate events were not necessary conditions for extreme socioeconomic disruptions and human impacts rendered by the colonial era famines in peninsular India. Based on our findings, we suggest that climate change risk assessement should consider the potential impacts of more frequent low-level anomalies (e.g. 1 SD) in drought prone semi-arid regions.

Bagging Machine Learning Algorithms: A Generic Computing Framework Based on Machine-Learning Methods for Regional Rainfall Forecasting in Upstate New York

Regional rainfall forecasting is an important issue in hydrology and meteorology. Machine learning algorithms especially deep learning methods have emerged as a part of prediction tools for regional rainfall forecasting. This paper aims to design and implement a generic computing framework that can assemble a variety of machine learning algorithms as computational engines for regional rainfall forecasting in Upstate New York. The algorithms that have been bagged in the computing framework include the classical algorithms and the state-of-the-art deep learning algorithms, such as K-Nearest Neighbors, Support Vector Machine, Deep Neural Network, Wide Neural Network, Deep and Wide Neural Network, Reservoir Computing, and Long Short Term Memory methods. Through the experimental results and the performance comparisons of these various engines, we have observed that the SVM- and KNN-based method are outstanding models over other models in classification while DWNN- and KNN-based methods outstrip other models in regression, particularly those prevailing deep-learning-based methods, for handling uncertain and complex climatic data for precipitation forecasting. Meanwhile, the normalization methods such as Z-score and Minmax are also integrated into the generic computing framework for the investigation and evaluation of their impacts on machine learning models.

Thermodynamics control precipitation in NE Mexico on orbital to millennial timescales

Northern Mexico is projected to become more arid in the future, however the magnitude, timing and spatial extent of precipitation change is presently poorly constrained. To address this, we have developed a multi-proxy (δ18O, δ13C, Mg/Ca) U-Th dated speleothem record of past rainfall variability spanning 4.6 to 58.5 ka from Tamaulipas, Mexico. Our results demonstrate a dominant thermodynamic control on hydroclimate via changes in Atlantic SSTs. Our record robustly demonstrates this response during major paleoclimate events including the Last Glacial Maximum, the Younger Dryas and Heinrich Stadials 1, 3, 4, and 5. While previous work has suggested the magnitude of the Caribbean Low-Level Jet as the predominant driver of regional rainfall, we utilize a state-of-the-art climate model to isolate cool Atlantic SSTs as the dominant mechanism of drying. We also demonstrate this response is consistent across large parts of Mesoamerica, suggesting drying in the future may be more spatially homogenous than currently predicted.

Prediction of Short-Time Rainfall Based on Deep Learning

Short-time heavy rainfall is a kind of sudden strong and heavy precipitation weather, which seriously threatens people’s life and property safety. Accurate precipitation nowcasting is of great significance for the government to make disaster prevention and mitigation decisions in time. In order to make high-resolution forecasts of regional rainfall, this paper proposes a convolutional 3D GRU (Conv3D-GRU) model to predict the future rainfall intensity over a relatively short period of time from the machine learning perspective. Firstly, the spatial features of radar echo maps with different heights are extracted by 3D convolution, and then, the radar echo maps on time series are coded and decoded by using GRU. Finally, the trained model is used to predict the radar echo maps in the next 1-2 hours. The experimental results show that the algorithm can effectively extract the temporal and spatial features of radar echo maps, reduce the error between the predicted value and the real value of rainfall, and improve the accuracy of short-term rainfall prediction.

Generalized linear models for analyzing count data of rainfall occurrences

Having the adequate knowledge about the behavior of climatic variables on the occurrences of rainfall is needed to the country’s economists and agriculturists for saving the country’s people from the devastating natural hazards like flash flood, drought, heavy rainfall, etc. Therefore, the study has been taken initiative to identify the influence of climatic variables for the occurrences of rainfall. The study has been developed generalized linear models (GLMs) for Poisson distribution for weekly and fortnightly count data of daily rainfall occurrences for the summer and monsoon seasons for five regional rainfall stations of Bangladesh. For these models, minimum and maximum temperatures and relative humidity are considered as explanatory variables. For five regional rainfall stations, the model selection procedures AIC and BIC indicate that the GLMs for the Poisson distribution satisfactorily explain the influence of climatic variables for the fortnightly occurrences of rainfall in the summer and monsoon seasons. The GLMs for the summer season of fortnightly occurrences of rainfall indicate that if one unit of relative humidity increases, then the probability of rainy days will be increased by 12 percent in Feni station, 6 percent in Sylhet, Khulna and Rajshahi stations, and 7 percent in Dhaka station. Besides, the GLMs for the monsoon season of fortnightly occurrences of rainfall indicate that if one unit increases of minimum temperature, then the probability of rainy days will be increased by 22 percent, 19 percent, 24 percent, 17 percent and 19 percent in Feni, Sylhet, Khulna, Rajshahi and Dhaka stations, respectively. Further, maximum temperature indicates negative influence on the occurrences of rainfall for all the stations and seasons of the period. The study indicates that the relative humidity for summer season and minimum temperature for monsoon season play remarkable role for changing fortnightly occurrences of rainfall in all the regions of the country.

Modeling the Interaction between Easterly Waves and Deep Convection in Costa Rica

<p>Easterly waves (EWs) are an important feature of the intertropical convergence zone, they often serve as precursors to tropical cyclones, and, during boreal summers, are one of the main contributors to rainfall in various countries in Central America. Given the land-sea configuration that features the region, a better understanding of the EWs impact on regional rainfall would leverage the comprehension of regional interactions processes. EWs were also one of the foci of OTREC (Organization of Tropical East Pacific Convection), an observational campaign that took place in Costa Rica and Colombia from 5 August to 9 October 2019. Here, we will present some results obtained with high-resolution numerical simulations conducted with the System for Atmospheric Modeling (SAM), which are based on data collected during OTREC. We will begin by presenting a series of simulations forced with high-frequency radiosonde data collected in Santa Cruz, Costa Rica, for a weeklong period during OTREC, highlighting model performance in reproducing the data. We will then discuss more idealized SAM simulations designed to investigate convective initiation and convective organization at various stages of EW passage. Finally, using sensitivity experiments with SAM in which we override soil moisture conditions, we will address the role of surface moisture in modulating the interaction between EWs and deep convection over land. This work aims to improve current knowledge on the role of EWs for regional rainfall, influence on the initiation of deep convection and further surface-atmosphere feedbacks.</p>