Hurricanes, rapidly increasing in complexity and strength in a warmer world, are one of the worst natural disasters in the 21st century. Further studies integrating the changing hurricane features are thus crucial to aid in the prediction of major hurricanes. With this in mind, we present a new framework based on automated decision tree analysis, which has the capability to identify the most important cloud structural parameters from GOES imagery as predictors for hurricane intensification potential in the Atlantic and Pacific oceans. The proposed framework has been proved effective for predicting major hurricanes with an overall accuracy of 73% from 6 to 54 h in advance (both regions combined).
Water quality is a subject of intense scientific inquiry because of its repercussion in human’s life, agriculture or even energy generation. Remote sensing can be used to control water masses by analyzing biophysical variables. Chlorophyll-a (Chl-a) and Total Suspended Solids (SS) are a well-known feature of water quality. These variables have been measured in Sitjar reservoir (Castelló, Spain) as a part of the project Ecological Status of Aquatic Systems with Sentinel Satellites (ESAQS), in order to compare the results with satellite reflectance data. Two processes were compared to correct atmospherically the level 1C Sentinel 2 (S2) images. The results show that Case 2 Regional Coast Colour (C2RCC) method, with a Root Mean Square Error of 2.4 mg/m3 (Chl-a) and 3.9 g/m3 (SS) is a better tool for atmospheric correction in this scenario due to the low turbidity levels of water. Besides, in this paper we study the Chl-a and SS variability through April 2017 to March 2019 with fourteen S2 images with the automatic products from C2RCC correction, finding correlations between them and the climate and reservoir conditions. Chl-a increase from 0.4 mg/m3 to 9.5 mg/m3 while SS rise 18 g/m3 in this period, which makes Sitjar as an oligotrophic-mesotrophic system. The correlation results demonstrate an excellent correspondence between them (R2=0.9). Sitjar reservoir lost almost 40 hm3 at the beginning of the study, which it had a possible relationship with the increasing parameter values. Also discussed was the role played by the climatology in the reservoir conditions due to the changes in the water structure with seasons, which explains the ariability through the year.
Regional-scale climate variability has far-reaching implications for both local and global climate systems, impacting factors like temperature, precipitation patterns, oceanic circulation, and the occurrence of extreme weather events. However, despite these influences, there is currently no universal methodology for the automated identification of regional-scale variability modes, including those less dominant globally, and for simultaneously exploring the influence of various ocean depth layers in characterizing these modes and diagnosing regional sea level variations. The presented innovative approach addresses these critical region-specific needs by assisting in the extraction of novel regional depth-layered variability modes and establishing their correlation with regional sea level fluctuations, employing tailored machine-learning techniques. This dual-purpose is achieved through the utilization of an optimized k-means clustering method for the automatic identification of regions with shared variability patterns across all global oceans, revealing previously unexplored regional variability modes. Additionally, guided by an EOF/PC analysis, the approach facilitates an automatic exploration of depth layers that significantly contribute to explaining sea level variability, providing insights into diverse climatic regions. Furthermore, the methodology is specifically designed for a multi-scale analysis, enabling the examination of climate variability spanning from months to several years. The results obtained through this approach have the potential to support informed decision-making regarding local climate-related changes.
The analysis of past regional climate‐related sea level variations has important implications for diagnosing changes in future sea level driven by climate fluctuations. As the climate changes, there is a need for new explanatory variables of within‐region climate factors and for more complex methods able to identify nonlinear relationships, such as machine learning algorithms. This study demonstrates the application of a new machine learning‐based methodology to reconstruct historical sea level tide gauge records from proxy data (i.e., upper‐ocean temperature estimates in open ocean regions), which provide a reasonably good dynamical representation of coastal sea level variations linked to slow and persistent natural processes like internal climate variability. The learning performance of our method was evaluated against observations of multiple stations and across a variety of model reconstructions, as shown and evidenced by the results.
Climate change and rising sea levels pose significant threats to coastal regions, necessitating accurate and timely forecasts. Current methods face limitations due to their inability to fully capture nonlinear complexities, high computational costs, gaps in historical data, and bridging the gap between short-term and long-term forecasting intervals. Our study addresses these challenges by combining advanced machine learning techniques to provide region-specific sea level predictions in the Mediterranean Sea. By integrating high-resolution sea surface temperature data spanning 40 years, we employed a tailored k-means clustering technique to identify regions of high variance. Using these clusters, we developed RNN-GRU models that integrate historical tide gauge data and sea surface height data, offering regional sea level predictions on timescales ranging from one month to three years. Our approach achieved the highest predictive accuracy, with correlation values ranging from 0.65 to 0.84 in regions with comprehensive datasets, demonstrating the model’s robustness. In areas with fewer tide gauge stations or shorter time series, our models still performed moderately well, with correlations between 0.51 and 0.70. However, prediction accuracy decreases in regions with complex geomorphology. Yet, all regional models effectively captured sea level variability and trends. This highlights the model’s versatility and capacity to adapt to different regional characteristics, making it invaluable for regional planning and adaptation strategies. Our methodology offers a powerful tool for identifying regions with similar variability and providing sub-regional scale predictions up to three years in advance, ensuring more reliable and actionable sea level forecasts for Mediterranean coastal communities.
Hurricanes, and more generally tropical cyclones, are among the most destructive natural hazards, and are arguably changing under climate change influences. Applying the power of AI to predict the extreme behavior of these events could be key to helping minimize hurricane damage. AI tools are a significant opportunity to: 1) identify non-linear relationships between changing hurricane-related characteristics and tropical storm intensification, and 2) anticipate responses to these changes. Another key part of this AI-based system is uncertainty quantification for decision-making processes. In this context, we present an improved ML hybrid model for predicting the development of extreme hurricane events, which includes effective information on spatio-temporal evolution variations of structural parameters extracted from IR satellite images. This approach, which combines Convolutional Neural Networks (CNNs) and a Random Forest (RF) classification framework, has been trained/tested with data from 1995 over the North Atlantic and NorthEast Pacific regions. Results from the CNN-RF model shows a performance of 80% or better considering lead-times of up to three days ahead (every 6 hours). With the proposed configuration, the overall precision has increased by at least 8%. This model could be yet further improved with the inclusion of new variables linked to environmental factors to be progressively explored.