Stennis Space Center

There are two primary mechanisms for carbon transfer into the geological record within the ocean: the organic sink and the inorganic carbon sink. The inorganic carbon sink (i.e., the long-term preservation of carbonate minerals in the geologic record), in particular, is not well constrained. Over the past several years, theoretical and experimental work has significantly improved our understanding of the inorganic carbon cycle, particularly in the context of carbonate minerals. This work reviews the state-of-the-art of marine calcium carbonate production, dissolution, and accumulation, representing a comprehensive update on the works of Milliman (1993) and Smith and Mackenzie (2016). We combine updated estimates with recent advances in understanding the carbonic acid system and the global ocean carbon cycle. These processes are supplemented by organismal to ecosystem-level responses of calcifying organisms to future climate change, enabling estimates of future global marine carbonate budget development. Our synthesis offers a comprehensive perspective on the long-term balance of carbonate preservation under current boundary conditions and projected future climate scenarios.Present-day net global annual gross production of CaCO3 throughout all carbonate-producing ecosystems is estimated between 176.1*1012 mol yr-1 (= 17.61 Gt yr-1) and 437.1*1012 mol yr-1 (= 437.1 Gt yr-1). However, only 8 to 20% of CaCO3 sediment is preserved each year, according to literature data. This leads to ~35*1012 mol yr-1 (= 3.5 Gt yr-1) of CaCO3 accumulation per year. Today, nearly all CaCO3 production is estimated to be biologically mediated or biologically controlled, while contributions from abiogenic carbonates are assumed to be minor. Dissolution of CaCO3 in the marine realm buffers excess CO2 from organic matter oxidation in the water column and sediment. The carbonate buffer effect within the carbonic acid system is expected to decline in future climate change scenarios, likely leading to more intense carbonate dissolution.Anthropogenic climate change, including excess CO2 in surface oceans, is further proposed to impact the CaCO3 production capabilities of all CaCO3-producing ecosystems in the ocean by affecting the diversity, abundance, and calcification potential of organisms. Predicting the future development of CaCO3 accumulation in the oceans is an evolving process. With the currently available data, estimating future CaCO3 accumulation is challenging due to multiple unknowns regarding disruptions and adaptations of the producers. The herein-reviewed data suggest that the global net CaCO3 accumulation will decrease by over 50 % to 19*1012 mol yr-1 (= 1.90 Gt yr-1) by 2050 AD, and further to 10.3*1012 mol yr-1 (= 1.03 Gt yr-1) by 2100 AD based on a “business as usual” CO2 emission model. ReferencesMilliman, J. D.: Production and accumulation of calcium carbonate in the ocean: Budget of a nonsteady state, Global Biogeochem Cy, 7, 927–957, https://doi.org/10.1029/93gb02524, 1993.Smith, S. V. and Mackenzie, F. T.: The Role of CaCO3 Reactions in the Contemporary Oceanic CO2 Cycle, Aquat Geochem, 22, 153–175, https://doi.org/10.1007/s10498-015-9282-y, 2016.

Using a deep neural network to detect a subpixel target in hyperspectral imaging is a difficult problem in remote sensing. The problem is caused by having only one full target pixel for training the target class and a weak target signal in subpixel target pixels for validation and having to solve a problem in high dimensional space resulting from a large number of spectral bands in hyperspectral imaging and a large number of weights in a deep neural network. The most recent algorithm is a deep neural network with two hidden layers, synthetic subpixel targets for training the target class, and variable stepsize for the gradient descent method and has been shown to be able to detect the subpixel target, but it is a high-dimensional problem resulting from setting the dimensions at the hidden layers to be the same as the number of spectral bands in order not to lose the weak target signal. In this paper an algorithm is developed by adding a hidden layer to the current neural network. Instead of using rectified linear unit, the dimensions at two of the hidden layers are set to be substantially lower than the spectral bands to reduce high computational cost and the potential problem with overfitting. Experimental results from two hyperspectral images, one with simulated subpixel target pixels for training and validating the algorithm and the other with simulated subpixel target pixels for training and actual subpixel target pixels for validation, have shown that the proposed algorithm can perform better than benchmark algorithms without losing the weak subpixel target signal.

At the 2021 United Nations (UN) Conference of the Parties (COP) meeting in Glasgow, Nigeria committed to renewable energy to contain carbon emissions responsible for global warming. Its subsequent national Energy Transition Plan pledged to cut emissions by phasing out fossil fuels. Decades of oil exploration in Nigeria’s Niger Delta have caused extensive environmental damage resulting in the destruction of traditional livelihoods, economic strangulation, and widespread poverty. Multinational oil corporations are now divesting rapidly, leaving behind a trail of environmental pollution without compensation to impacted communities. The clash between fossil-fuel phase-outs and local people’s environmental justice demands demonstrate the imperative of embedding justice and inclusion into green energy solutions. Drawing from the experiences of Nigeria’s oil-rich communities in the Niger Delta, this article argues that placing justice at the centre of national and global decarbonisation agendas is essential for ensuring that greener solutions do not contribute to human rights abuses and exacerbate pre-existing environmental challenges.

Flow cytometry typically reveals two heterotrophic prokaryote (HP) subpopulations when stained with SYBR Green: high nucleic acid (HNA) and low nucleic acid (LNA) cells. Evidence suggests these populations have distinct physiological and ecological roles with implications for mortality. We assessed HP abundance, production, the relative proportion of HNA and LNA, virus-mediated mortality, and the distribution of lytic versus lysogenic strategies within HP host communities across a latitudinal gradient in the North Atlantic during spring. The study area, characterized by dynamic physicochemical conditions consistent with the onset of seasonal stratification, was divided into three regions based on bio-physicochemical properties: Pre-bloom, Bloom, and Oligotrophic. Multivariant analysis showed these regions significantly structured HPs, as well as influenced the relative abundance and production of virus subpopulations (i.e., V1 and V2). Specifically, V1 viruses increased with the potential of encountering HNA hosts, which were elevated in the surface waters of stratified Oligotrophic and Bloom regions. In contrast, V2 abundance and production correlated with LNA cells, more prominent in DEEP samples and in surface waters of the deeper mixed Pre-bloom region. Lysogeny occurred across all regions, with the percentage of lysogens within the HP community, increasing (largely V1-driven) with HP-specific growth rate until reaching a threshold of 0.1 d−1, after which it declined. We discuss the potential ecological underpinnings driving these patterns and implications for carbon flux.

When forecasting fixed-location observation nodes with statistical surrogate models, combining datasets during training loss calculation has been shown to improve model accuracy. However, traditional methods for tuning the data ratio, such as grid sea

In this work, we describe the distribution and variation of subsurface acoustic ducts formed by North Atlantic Subtropical Mode Water in the region where the Gulf Stream encounters the New England Seamount chain. High-resolution observations were obtained during two cruises (November 2022 and May/June 2023) from ships and teams of up to seven gliders as part of the Predictions of AcousticS with Smart Experimental Networks of GlidERS (PASSENGERS) project. Low stratification mode water layers of varying thickness were widely observed from gliders, shipboard CTD, and towed underway CTD casts. The cruises bracket the wintertime formation of the mode waters and are located near the region of maximum production. Variation in the mode water properties on the scale of the local baroclinic Rossby radius was observed resulting from an eddy field associated with the southern flank of the Gulf Stream. A strong streamwise/spanwise asymmetry in the ducts exists in the region. The seasonal production and erosion of mode water south of the Gulf Stream extension imprints on the subsurface duct, with thicker ducts and lower sound speed minima in the spring relative to the fall.

In this article, we introduce a novel framework for learning spatial concepts within a human-in-the-loop (HITL) context, highlighting the critical role of explainability in AI systems. By incorporating human feedback, the approach enhances the learning process, making it particularly suitable for applications where user trust and interpretability are essential, such as AiTR. Namely, we introduce a new parametric similarity measure for spatial relations expressed as histograms of forces (HoFs). Next, a spatial concept is represented as a spatially attributed graph and HoF bundles. Last, a process is outlined for utilizing this structure to make decisions and learn from human feedback. The framework’s robustness is demonstrated through examples with diverse user types, showcasing how varying feedback strategies influence learning efficiency, accuracy, and ability to tailor the system to a particular user. Overall, this framework represents a promising step toward human-centered AI systems capable of understanding complex spatial relationships while offering transparent insights into their reasoning processes.