Noveltis (France)

Sea surface salinity and temperature are essential climate variables in monitoring and modeling ocean health. Multispectral ocean color satellites allow the estimation of these properties at a resolution of 10 to 300 m, which is required to correctly represent their spatial variability in coastal waters. This paper investigates the effect of pre-applying an unsupervised classification in the performance of both temperature and salinity inversion. Two methodologies were explored: clustering based solely on spectral radiances, and clustering applied directly to satellite images. The former improved model generalization by identifying similar water clusters across different locations, reducing location dependency. It also demonstrated results correlating cluster type with salinity and temperature distributions thereby enhancing regression model performance and improving a global ocean color sea surface temperature regression model RMSE error by 10%. The latter approach, applying clustering directly to satellite images, incorporated spatial information into the models and enabled the identification of front boundaries and gradient information, improving global sea surface temperature models RMSE by 20% and sea surface salinity models by 30%, compared to the initial ocean color model. Beyond improving algorithm performance, optical water classification can be used to monitor and interpret changes to water optics, including algal blooms, sediment disturbance or other climate change or antropogenic disturbances. For example, the clusters have been used to show the impact of a category 4 hurricane landfall on the Mississippi estuarine region.

Sea surface salinity and temperature are essential climate variables in monitoring and modeling ocean health. Multispectral ocean color satellites allow the estimation of these properties at a resolution of 10 to 300 m, which is required to correctly represent their spatial variability in coastal waters. This paper investigates the effect of pre-applying an unsupervised classification in the performance of both temperature and salinity inversion. Two methodologies were explored: clustering based solely on spectral radiances, and clustering applied directly to satellite images. The former improved model generalization by identifying similar water clusters across different locations, reducing location dependency. It also demonstrated results correlating cluster type with salinity and temperature distributions thereby enhancing regression model performance and improving a global ocean color sea surface temperature regression model RMSE error by 10%. The latter approach, applying clustering directly to satellite images, incorporated spatial information into the models and enabled the identification of front boundaries and gradient information, improving global sea surface temperature models RMSE by 20% and sea surface salinity models by 30%, compared to the initial ocean color model. Beyond improving algorithm performance, optical water classification can be used to monitor and interpret changes to water optics, including algal blooms, sediment disturbance or other climate change or antropogenic disturbances. For example, the clusters have been used to show the impact of a category 4 hurricane landfall on the Mississippi estuarine region.

Sea surface salinity and temperature are essential climate variables in monitoring and modeling ocean health. Multispectral ocean color satellites allow the estimation of these properties at a resolution of 10 to 300 m, which is required to correctly represent their spatial variability in coastal waters. This paper investigates the effect of pre-applying an unsupervised classification in the performance of both temperature and salinity inversion. Two methodologies were explored: clustering based solely on spectral radiances, and clustering applied directly to satellite images. The former improved model generalization by identifying similar water clusters across different locations, reducing location dependency. It also demonstrated results correlating cluster type with salinity and temperature distributions thereby enhancing regression model performance and improving a global ocean color sea surface temperature regression model RMSE error by 10%. The latter approach, applying clustering directly to satellite images, incorporated spatial information into the models and enabled the identification of front boundaries and gradient information, improving global sea surface temperature models RMSE by 20% and sea surface salinity models by 30%, compared to the initial ocean color model. Beyond improving algorithm performance, optical water classification can be used to monitor and interpret changes to water optics, including algal blooms, sediment disturbance or other climate change or antropogenic disturbances. For example, the clusters have been used to show the impact of a category 4 hurricane landfall on the Mississippi estuarine region.

As cloud adoption accelerates across enterprise environments, organizations increasingly face the complexity of managing large-scale, distributed systems spanning multiple regions and accounts. The challenge of maintaining effective monitoring and alerting mechanisms across such vast infrastructures has become paramount for ensuring system reliability, security, and performance. This paper presents a comprehensive investigation into the implementation of centralized alerting frameworks within cloud environments, with particular emphasis on Amazon Web Services (AWS) cloud-native tools including CloudWatch, Simple Notification Service (SNS), and Lambda, alongside third-party monitoring solutions such as New Relic and Splunk. The centralized approach to alerting addresses the inherent complexities of distributed cloud architectures by consolidating alert management, reducing operational overhead, and improving response times to critical system events. Through systematic analysis of architectural patterns, implementation strategies, and best practices, this research demonstrates how organizations can achieve scalable, resilient, and cost-effective alerting solutions. The paper examines multi-region deployment strategies, distributed processing mechanisms, and high-availability patterns that ensure continuous monitoring capabilities even during regional service disruptions. Performance evaluation and case study analysis reveal significant improvements in mean time to detection (MTTD) and mean time to resolution (MTTR) when compared to traditional decentralized alerting approaches.

Abstract. Eurasian rivers provide a quarter of total fresh water to the Arctic, maintaining a persistent fresh layer that covers the surface Arctic Ocean. This freshwater export controls Arctic Ocean stratification, circulation, and basin-wide sea ice concentration. The Lena River supplies the largest volume of runoff and plays a key role in this system, as runoff outflows into the Laptev Sea as a particularly shallow plume. Previous in situ and modelling studies suggest that local wind forcing is a driver of variability in Laptev sea surface salinity (SSS) but there is no consensus on the roles of Lena River discharge and sea ice cover in contributing to this variability or on the dominant driver of variability. Until recently, satellite SSS retrievals were insufficiently accurate for use in the Arctic. However, retreating sea ice cover and continuous progress in satellite product development have significantly improved SSS retrievals, giving satellite SSS data true potential in the Arctic. In this region, satellite-based SSS is found to agree well with in situ data (r>0.8) and provides notable improvements compared to the reanalysis product used in this study (r>0.7) in capturing patterns and variability observed in in situ data. This study demonstrates a novel method of identifying the dominant drivers of interannual variability in Laptev Sea dynamics within reanalysis products and testing if these relationships appear to hold in satellite-based SSS, sea surface temperature (SST) data, and in situ observations. The satellite SSS data firmly establish what is suggested by reanalysis products and what has previously been subject to debate due to the limited years and locations analysed with in situ data; the zonal wind is the dominant driver of offshore or onshore Lena River plume transport. The eastward wind confines the plume to the southern Laptev Sea and drives alongshore transport into the East Siberian Sea, and westward wind drives offshore plume transport into the northern Laptev Sea. This finding is affirmed by the strong agreement in SSS pattern under eastward and westward wind regimes in all reanalyses and satellite products used in this study, as well as with in situ data. The pattern of SST also varies with the zonal wind component and drives spatial variability in sea ice concentration.

Abstract In the crucial region of the Yermak Plateau where warm Atlantic water enters the Arctic ocean, we examined high frequency variations in the Yermak Pass Branch over a 34 months‐long mooring data set. The mooring was ice covered only half of the time with ice‐free periods both in summer and winter. We investigated the contribution of residual tidal currents to the low frequency flow of Atlantic Water (AW) and high frequency variations in velocity shears possibly associated with internal waves. High resolution model simulations including tides show that diurnal tide forced an anticyclonic circulation around the Yermak Plateau. This residual circulation helps the northward penetration of the AW into the Arctic. Tides should be taken into account when examining low frequency AW inflow. High frequency variations in velocity shears are mainly concentrated in a broad band around 12 hr in the Yermak Pass. Anticyclonic eddies, observed during ice‐free conditions, modulate the shear signal. Semi‐diurnal internal stationary waves dominate high frequency variations in velocity shears. The stationary waves could result from the interaction of freely propagating semi‐diurnal internal waves generated by diurnal barotropic tides on critical slopes around the plateau. The breaking of the stationary waves with short length scales possibly contribute to mixing of AW at the entrance to the Arctic.

<strong class="journal-contentHeaderColor">Abstract.</strong> The flow of carbon through terrestrial ecosystems and the response to climate are critical but highly uncertain processes in the global carbon cycle. However, with a rapidly expanding array of in situ and satellite data, there is an opportunity to improve our mechanistic understanding of the carbon (C) cycle's response to land use and climate change. Uncertainty in temperature limitation on productivity poses a significant challenge to predicting the response of ecosystem carbon fluxes to a changing climate. Here we diagnose and quantitatively resolve environmental limitations on the growing-season onset of gross primary production (GPP) using nearly 2 decades of meteorological and C flux data (2000–2018) at a subalpine evergreen forest in Colorado, USA. We implement the CARbon DAta-MOdel fraMework (CARDAMOM) model–data fusion network to resolve the temperature sensitivity of spring GPP. To capture a GPP temperature limitation – a critical component of the integrated sensitivity of GPP to temperature – we introduced a cold-temperature scaling function in CARDAMOM to regulate photosynthetic productivity. We found that GPP was gradually inhibited at temperatures below 6.0 <span class="inline-formula">^∘</span>C (<span class="inline-formula">±</span>2.6 <span class="inline-formula">^∘</span>C) and completely inhibited below <span class="inline-formula">−</span>7.1 <span class="inline-formula">^∘</span>C (<span class="inline-formula">±</span>1.1 <span class="inline-formula">^∘</span>C). The addition of this scaling factor improved the model's ability to replicate spring GPP at interannual and decadal timescales (<span class="inline-formula"><i>r</i>=0.88</span>), relative to the nominal CARDAMOM configuration (<span class="inline-formula"><i>r</i>=0.47</span>), and improved spring GPP model predictability outside of the data assimilation training period (<span class="inline-formula"><i>r</i>=0.88</span>). While cold-temperature limitation has an important influence on spring GPP, it does not have a significant impact on integrated growing-season GPP, revealing that other environmental controls, such as precipitation, play a more important role in annual productivity. This study highlights growing-season onset temperature as a key limiting factor for spring growth in winter-dormant evergreen forests, which is critical in understanding future responses to climate change.

Robotic process automation (RPA) is one of the most emerging technology areas of the last decade. As the name implies, RPA is an approach to automate repetitive tasks in business operations. Many solutions are avail-able on the market by multiple vendors. Through the implementation of those RPA solutions, companies can achieve higher performance levels and lead a differentiating competitive edge. One of the first fields which have benefited from RPA is Supply Chain. This paper presents a solution for the task selection problem issue of RPA applied to the Supply Chain. A case study is also presented to demonstrate the effectiveness of the designed solution.

Using Natural Language (NL) to interacting with relational databases allows users from any background to easily query and analyze large amounts of data. This requires a system that understands user questions and automatically converts them into structured query language such as SQL. The best performing Text-to-SQL systems use supervised learning (usually formulated as a classification problem) by approaching this task as a sketch-based slot-filling problem, or by first converting questions into an Intermediate Logical Form (ILF) then convert it to the corresponding SQL query. However, non-supervised modeling that directly converts questions to SQL queries has proven more difficult. In this sense, we propose an approach to directly translate NL questions into SQL statements. In this study, we present a Sequence-to-Sequence (Seq2Seq) parsing model for the NL to SQL task, powered by the Transformers Architecture exploring the two Language Models (LM): Text-To-Text Transfer Transformer (T5) and the Multilingual pre-trained Text-To-Text Transformer (mT5). Besides, we adopt the transformation-based learning algorithm to update the aggregation predictions based on association rules. The resulting model achieves a new state-of-the-art on the WikiSQL DataSet, for the weakly supervised SQL generation.