Bayesian and Neural Network Approaches to Estimate Deep Temperature Distribution for Assessing a Supercritical Geothermal System: Evaluation Using a Numerical Model

The temperature distribution at depth is a key variable when assessing the potential of a supercritical geothermal resource as well as a conventional geothermal resource. Data-driven estimation by a machine-learning approach is a promising way to estimate temperature distributions at depth in geothermal fields. In this study, we developed two methodologies—one based on Bayesian estimation and the other on neural networks—to estimate temperature distributions in geothermal fields. These methodologies can be used to supplement existing temperature logs, by estimating temperature distributions in unexplored regions of the subsurface, based on electrical resistivity data, observed geological/mineralogical boundaries, and microseismic observations. We evaluated the accuracy and characteristics of these methodologies using a numerical model of the Kakkonda geothermal field, Japan, where a temperature above 500 °C was observed below a depth of about 3.7 km. When using geological and geophysical knowledge as prior information for the machine learning methods, the results demonstrate that the approaches can provide subsurface temperature estimates that are consistent with the temperature distribution given by the numerical model. Using a numerical model as a benchmark helps to understand the characteristics of the machine learning approaches and may help to identify ways of improving these methods.

     

Attenuation behaviour of tuffaceous sandstone and granite during microfracturing

Laboratory measurements of ultrasonic wave propagation in tuffaceous sandstone (Kimachi, Japan) and granite (Iidate, Japan) were performed during increasing fracturing of the samples. The fracturing was achieved by unconfined uniaxial compression up to and beyond the point of macrofracture of the specimen using a constant low strain rate. The observed variation of wave velocity (up to 40 per cent) due to the development of micro- and macrofractures in the rock is interpreted by rock models relating velocity changes to damage and crack density. The calculated density of the newly formed cracks reaches higher values for the sandstone than for the granite. Using the estimated crack densities, the attenuation behaviour is interpreted in terms of different attenuation mechanisms; that is, friction and scattering. Rayleigh scattering as described by the model of Hudson (1981 ) may explain the attenuation qualitatively if the largest plausible crack dimensions are assumed in modelling.     

Calibration and validation of the ocean color version-3 product from ADEOS OCTS

We present calibration and validation results of the OCTS’s ocean color version-3 product, which mainly consists of the chlorophyll-a concentration (Chl-a) and the normalized water-leaving radiance (_nL_w). First, OCTS was calibrated for the inter-detector sensitivity difference, offset, and absolute sensitivity using external calibration source. It was also vicariously calibrated using in-situ measurements for water (Chl-a and_nL_w) and atmosphere (optical thickness), which were acquired synchronously with OCTS under cloud-free conditions. Second, the product was validated using selected 17 in-situ Chl-a and 11 in-situ_nL_w measurements. We confirmed that Chl-a was estimated with an accuracy of 68% for Chl-a less than 2 mg/m³, and_nL_w from 94% (band 2) to 128% (band 4). Geometric accuracy was improved to 1.3 km. Stripes were significantly reduced by modifying the detector normalization factor as a function of input radiance.     

Correction of atmospheric effect on ADEOS/OCTS ocean color data: Algorithm description and evaluation of its performance

This paper first describes the atmospheric correction algorithm for OCTS visible band data used at NASDA/EOC. Sharing a basic structure with Gordon and Wang’s Sea WiFS algorithm, it uses 10 candidate aerosol models including the “Asian dust model” introduced in consideration of the unique feature of aerosols over the east Asian waters. Based on the observations at 670 and 865 nm bands, the algorithm selects a pair of aerosol models that account best for the observed spectral reflectances, and synthesizes the aerosol reflectance used for the atmospheric correction. Two different schemes for determining the value of the parameter for the aerosol model selection are presented and their anticipated estimation error is analyzed in terms of retrieved water reflectance at 443 nm. The results of our numerical simulation show that the standard deviation of the estimation error of the “weighted average” scheme is mostly within the permissible level of ±0.002, reducing the error by 18% on average compared to the “simple average” scheme. The paper further discusses the expected error under the old CZCS-type atmospheric correction, which assumes constant aerosol optical properties throughout the given image. Although our algorithm has a better performance than the CZCS algorithm, further analysis shows that the error induced by the assumption taken in the algorithm that the water-leaving radiance at 670 nm band is negligibly small may be large in high pigment concentration waters, indicating the necessity for future improvements.     

Improved Method for Calculating Anthropogenic CO2 in the Upper Layer of the North Pacific Subtropical Gyre along 175°E

Seawater samples were collected in the North Pacific along 175°E during a cruise of the Northwest Pacific Carbon Cycle Study (NOPACCS) program in 1994. Many properties related to the carbonate system were analyzed. By using well-known ratios to correct for chemical changes in seawater, the CO₂ concentration at a given depth was back calculated to its initial concentration at the time when the water left the surface in winter. We estimated sea-surface CO₂ and titration alkalinity (TA) in present-day winter, from which we evaluated the degree of air-sea CO₂ disequilibrium in winter was. Using a correction factor for air-sea CO₂ disequilibrium in winter, we reconstructed sea-surface CO₂ in pre-industrial times. The difference between the back-calculated initial CO₂ and sea-surface CO₂ in pre-industrial times should correspond to anthropgenic CO₂ input. Although the mixing of different water masses may cause systematic error in the calculation, we found that the nonlinear effect induced by the mixing of different water masses was negligible in the upper layer of the North Pacific subtropical gyre along 175°E. The results of our improved method of assessing the distribution of anthropogenic CO₂ in that region show marked differences from those obtained using the previous back-calculation method.     

Improving integrity of RC beam-column joints with deficient beam rebar anchorage

In recent earthquakes in developing countries, severe damage was observed on reinforced concrete buildings. This study focuses on exterior beam-column joints with substandard beam rebar anchorage and seismic strengthening by installing wing walls. First, a series of experiments was conducted to investigate the seismic behavior of exterior joints with substandard beam rebar anchorage representing typical Bangladeshi buildings. Two 0.7-scaled exterior joint specimens were tested, and these specimens showed beam rebar anchorage failure and/or joint shear failure. Prior to strengthening of the joint, a series of pullout tests was conducted on postinstalled bonded anchors in low-strength concrete for strengthening design. Then, an experiment was performed to apply the strengthening method by wing walls to one of the exterior joint specimens to improve the integrity, and this method was intended to prevent the failure of beam rebar anchorage. The strengthening method is proposed to extend the development length of beam longitudinal bars by considering the embedment length along the wing walls. The test results verified the effectiveness and applicability of the proposed strengthening method to upgrade exterior RC beam-column joints with deficient beam rebar anchorage.     

The Ganges and Brahmaputra rivers in Bangladesh: basin denudation and sedimentation

Every year the Ganges and Brahmaputra rivers in Bangladesh transport 316 and 721 million tonnes of sediment, respectively. These high loads of suspended sediment reflect the very high rate of denudation in their drainage basins. The average mechanical denudation rate for the Ganges and Brahmaputra basins together is 365 mm 10³ yr⁻¹. However, the rate is higher in the Brahmaputra Basin than that in the Ganges Basin. Several factors, including mean trunk channel gradient, relief ratio, runoff, basin lithology and recurring earthquakes are responsible for these high denudation rates. Of the total suspended sediment load (i.e. 1037 million tonnes) transported by these rivers, only 525 million tonnes (c. 51% of the total load) are delivered to the coastal area of Bangladesh and the remaining 512 million tonnes are deposited within the lower basin, offsetting the subsidence. Of the deposited load, about 289 million tonnes (about 28% of the total load) are deposited on the floodplains of these rivers. The remaining 223 million tonnes (about 21% of the total load) are deposited within the river channels, resulting in aggradation of the channel bed at an average rate of about 3·9 cm yr⁻¹. Although the Brahmaputra transports a higher sediment load than the Ganges, the channel bed aggradation rate is much higher for the Ganges. This study also documents a wide range of interannual, seasonal and daily variation in suspended sediment transport and water discharge. Interannual variation in sediment deposition within the basin is also suggested. Copyright © 1999 John Wiley & Sons, Ltd.     

Remote effects of mixed layer development on ocean acidification in the subsurface layers of the North Pacific

Using the outputs of projections under the highest emission scenario of the representative concentration pathways performed by Earth system models (ESMs), we evaluate the ocean acidification rates of subsurface layers of the western North Pacific, where the strongest sink of atmospheric CO₂ is found in the mid-latitudes. The low potential vorticity water mass called the North Pacific Subtropical Mode Water (STMW) shows large dissolved inorganic carbon (DIC) concentration increase, and is advected southwestward, so that, in the sea to the south of Japan, DIC concentration increases and ocean acidification occurs faster than in adjacent regions. In the STMW of the Izu-Ogasawara region, the ocean acidification occurs with a pH decrease of ~0.004 year^?1 , a much higher rate than the previously estimated global average (0.0023 year^?1), so that the pH decreases by 0.3–0.4 during the twenty-first century and the saturation state of calcite (Ω_Ca) decreases from ~4.8 down to ~2.4. We find that the ESMs with a deeper mixed layer in the Kuroshio Extension region show a larger increase in DIC concentration within the Izu-Ogasawara region and within the Ryukyu Islands region. Comparing model results with the mixed layer depth obtained from the Argo dataset, we estimate that DIC concentration at a depth of ~200 m increases by 1.4–1.6 μmol kg^?1 year^?1 in the Izu-Ogasawara region and by 1.1–1.4 μmol kg^?1 year^?1 in the Ryukyu Islands region toward the end of this century.     

Nutrients,light and phytoplankton production in the shallow,tropical coastal waters of Bandon Bay,Southern Thailand

Phytoplankton primary production and its regulation by light and nutrient availability were investigated in the shallow, tropical coastal waters of Bandon Bay, Southern Thailand. The bay was meso‐eutrophicated and highly turbid, receiving river water discharge. Water column stratification was consistently weak during both rainy and dry seasons. Dissolved inorganic nitrogen (DIN) was higher off the river mouth than in the other regions, suggesting that river water discharge was a main source of DIN. By contrast, dissolved inorganic phosphorus (DIP) showed a significant negative correlation with total water depth, implying that regeneration around the sea floor was an important source of DIP. Surface DIN and DIP showed positive correlations with surface primary production (PP) and water column primary productivity (ΣPP*), respectively. The combined correlation and model analyses indicate that total water depth had an ambivalent influence on water column primary production (ΣPP); shallower water depth induced more active regeneration of nutrients, but it also caused higher turbidity and lower light availability as a result of enhanced resuspension of sediments. Furthermore, there was a vertical constraint for phytoplankton during the rainy season: total water depth tended to be shallower than euphotic zone depth. In conclusion, light limitation and vertical constraint owing to shallow water depth appear to be more important than nutrient limitation for water column primary production in Bandon Bay.