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.
Volatile organic iodine compounds (VOIs) emitted from the ocean surface to the air play an important role in atmospheric chemistry. Shipboard observations were conducted in Funka Bay, Hokkaido, Japan, bimonthly or monthly from March 2012 to December 2014, to elucidate the seasonal variations of VOI concentrations in seawater and their sea-to-air iodine fluxes. The bay water exchanges with the open ocean water of the North Pacific twice a year (early spring and autumn). Vertical profiles of CH2I2, CH2ClI, CH3I, and C2H5I concentrations in the bay water were measured bimonthly or monthly within an identified water mass. The VOI concentrations began to increase after early April at the end of the diatom spring bloom, and represented substantial peaks in June or July. The temporal variation of the C2H5I profile, which showed a distinct peak in the bottom layer from April to July, was similar to the PO43? variation profile. Correlation between C2H5I and PO43? concentrations (r = 0.93) suggests that C2H5I production was associated with degradation of organic matter deposited on the bottom after the spring bloom. CH2I2 and CH2ClI concentrations increased substantially in the surface and subsurface layers (0–60 m) in June or July resulted in a clear seasonal variation of the sea-to-air iodine flux of the VOIs (high in summer or autumn and low in spring). 相似文献
A sediment trap experiment was carried out in conjunction with an over flight of Ocean Color Temperature Scanner (OCTS) on
board Advanced Earth Observing Satellite (ADEOS) at 40°N, 143°E off Sanriku in April to May 1997. Short term variability of
particle fluxes was examined at depths of 450 m and 600 m from April 6 to May 1 with a sampling interval of two days, and
at 450 m with one day interval from 2nd to 10th May. Daily averaged mass flux at 450 m and 600 m was 815 mg m−2d−1 and 862 mg m−2d−1, respectively. A sharp increase in mass flux was observed during the period from April 26 to April 29 with the highest mass
flux of 8 g m−2d−1. About 85% of the total mass flux for the entire duration (26 days) was collected within these 4 days. Trapped material during
the peak flux period was mainly composed of diatoms dominated byThalassiosira spp. and resting spores ofChaetoceros spp. This suggested that the peak flux was the result of (a) diatom bloom(s) in the euphotic column. Current meter records
at 420 m showed that on April 26 and 27, the period when the peak flux was observed, the southwestward current had diminished
in strength and changed its direction northwestward. Low current speeds appeared to have enhanced trap efficiency to help
form the peak flux. A time series of OCTS Intensive-LAC (Local Area Coverage: Region B) images from mid-March to early May
was examined todetect phytoplankton bloom(s). In the March 26th Chl image, high concentration region was restricted to the
southwest off Cape Erimo, but spread around the warm core ring (WCR) 93A by April 10. East of the WCR93A, high Chl concentration
remained steady until May, but to the west of the WCR93A, Chl decreased rapidly before the 19th of April. From this observation
we suspect that the peak flux observed at the end of April originated from a bloom, which ceased on the 17th or 18th of April,
in the region north of 40°N and west of 143°E. Taking the current meter records into account, the source region for the trapped
material is most likely around southwest of the Cape Erimo. 相似文献
The variety in shape and magnitude of thein vivo chlorophyll-specific absorption spectra of phytoplankton was investigated in relation to differences in pigment composition
off Sanriku, northwestern North Pacific. Site-to-site variations of the absorption coefficients,aph* (λ), and pigment composition were clearly observed. At warm-streamer stations, higher values ofaph* (440) andaph* (650) were found with relatively high concentrations of chlorophyllb (a green algae marker). At stations located in the Oyashio water (cold streamer),aph* (440) values were lower and fucoxanthin (a diatom marker) concentrations were higher, compared to the other stations. The
peak in the absorption spectra at the Oyashio stations was shifted toward shorter wavelengths, which was probably due to the
presence of phaeopigments. In a Kuroshio warm-core ring, the magnitude ofaph* (440) was in between those at the warm-streamer and Oyashio stations, and the diagnostic pigment was peridinin (a dinoflagellate
marker). These findings indicated that major differences in phytoplankton absorption spectra of each water mass were a result
of differences in the phytoplankton pigment composition of each water mass, which was probably related to the phytoplankton
community. 相似文献
—We investigated the effects of various viscoelastic structures on postseismic surface displacement and principal strain fields associated with the great 1946 Nankaido earthquake, which occurred on the plate boundary between the subducting Philippine Sea plate and the continental Eurasian plate. For this purpose, we constructed two kinds of three-dimensional structural models using the finite element method one is the Layered Model, in which a semi-infinite Maxwell viscoelastic material is underlying an elastic layer, and the other is the more realistic Plate Model, in which the three-dimensional configuration of the subducted Philippine Sea plate is taken into account. We also considered two cases with different thicknesses of the elastic layer (50 and 33km) for the respective models. The difference between the two models in postseismic surface deformations is significant for the case with the thinner elastic layer. In this case the horizontal surface displacement and principal strain for the Layered Model is two to three times larger than those for the Plate Model. Downward surface deformation tends to be dominant for the Layered Model, while the change in the pattern for the Plate Model is less marked. The spatial extent of uplift and subsidence for the Plate Model is broader than that for the Layered Model. Postseismic vertical displacements in Shikoku were found to be strongly dependent on the viscoelastic structures. From these results, we suggest that the estimates of the viscosity of the uppermost mantle, interplate coupling, and the area and the amount of after-slip following the 1946 Nankaido earthquake, which have been estimated based on simple layered viscoelastic models, should be re-evaluated using realistic three-dimensionally heterogeneous viscoelastic structures. 相似文献
Urea is an unstable and intermediate organic nitrogenous compound present in coastal environments and is derived from the excretion of some aquatic organisms and wastewater discharges. Urea plays an important role in the nitrogen cycle, where it is utilized by algae, including diatoms. However, there are very limited relevant data on the production, consumption, and degradation of urea because of the lack of appropriate measurement techniques. The conventional method is based on the formation of a colored product when urea reacts with diacetyl monoxime in a sulfuric acid solution. We examined the optimal conditions for the formation of the colored product; specifically, we evaluated different temperatures (22–80 °C), reaction times, mixing ratios of color reagents, and sample storage times. Application of the single mixed color-developing reagent (COLDER) at 70 °C resulted in the optimal formation of the colored product within a short reaction time of 60 min. This method was then used to measure dissolved urea in different coastal environments. The concentrations detected were as follows: 0.65–0.72, 0.49–0.58, and 1.09–2.28 µM urea-N at coral reef, seagrass, and mangrove sites, respectively. Our results showed high precision (SD = 0.02, CV = 1.2%), a low detection limit (0.03 µM urea-N), and a high recovery rate (94–99%). In summary, this high-temperature procedure for urea measurements should be valuable for obtaining high-precision data that can further the understanding of urea dynamics and its role in coastal ecosystems. 相似文献
The influence of organic ligands on natural phytoplankton growth was investigated in high-nitrate low-chlorophyll (HNLC) waters and during a phytoplankton bloom induced by a mesoscale iron enrichment experiment (SEEDS II) in the western subarctic Pacific. The growth responses of the phytoplankton in the treatments with iron complexed with model ligand were compared with those with inorganic iron or a control. Desferrioxamine B and protoporphyrin IX were used as models for hydroxamate-type siderophore and tetrapyrrole-type cell breakdown ligand, respectively. In the HNLC water, iron associated with protoporphyrin IX especially stimulated smaller phytoplankton (<10 μm) growth, 1.5-fold more than did inorganic iron. Surprisingly, only the addition of protoporphyrin IX stimulated small phytoplankton growth, suggesting that these cell breakdown ligands might be more bioavailable for them. The protoporphyrin IX’s stimulatory effect on small phytoplankton was not observed during bloom decline phase. The growth of phytoplankton was inhibited in the treatment with desferrioxamine B-complexed iron, suggesting its low bioavailability for the natural phytoplankton community. Its inhibitory effects were particularly pronounced in pico-eukaryotic phytoplankton. During the iron-induced bloom, the phytoplankton’s iron-stress response gradually increased with the desferrioxamine B concentration, suggesting that the competition for iron complexation between natural ligands and desferrioxamine B affected phytoplankton growth. However, the pico-eukaryotes did seem better able to utilize the desferrioxamine B-complexed iron during the bloom-developing phase. These results indicate that the iron bioavailability for phytoplankton differs between bloom-developing and bloom-decline phases. 相似文献
The inner part of the Ariake Sea is one of the most productive estuarine systems in Japan. To examine potential food items for estuarine organisms, we conducted monthly observations of the dynamics of particulate organic matter along the macrotidal Chikugo River estuary in 2005 and 2006. In the neighboring macrotidal Midori and Kuma River estuaries, comparative observations were made. High turbidity and strong vertical mixing were observed only at low salinities (<10) in the Chikugo River estuary. In contrast, the Midori and Kuma River estuaries were characterized by less turbid and less mixed waters. Concentrations of particulate organic carbon often exceeded 5?mg?l?1 in or close to the estuarine turbidity maximum (ETM) of the Chikugo River estuary. However, such high concentrations were rarely observed in the other two estuaries. The observed differences could be attributable to different hydrodynamic processes related to the different lengths of tidal reaches: 23, 8, and 6?km in the Chikugo, Midori, and Kuma Rivers, respectively. In the Chikugo River estuary, spatiotemporal changes of chlorophyll a suggested that phytoplankton occurred abundantly up- and/or downstream from the ETM especially during the warm season. In contrast, pheophytin (i.e., plant detritus) always accumulated in or close to the ETM. Carbon stable isotope ratios and carbon to nitrogen ratios indicated that the plant detritus was derived from phytoplankton and terrestrial plants. The Chikugo River estuary has a high potential to support the production of estuarine organisms through abundant plant detritus in the well-developed ETM all the year round. 相似文献
Spectroscopic observations of OH airglow undertaken on May 2, 2006 at Uji, Japan reveal variations in intensity and rotational temperature related to the passage of an atmospheric gravity wave. The variations exhibit a period of approximately 1 h and magnitudes of 2–6% in intensity and 0.5–2% in rotational temperature. The vertical wavelength and intrinsic frequency of the atmospheric gravity wave were determined from the horizontal wavelength derived by an OH airglow imager, the background horizontal wind velocity obtained by the middle and upper atmosphere (MU) radar, and the dispersion relationship. The observed variations are consistent with the values calculated using the model of Liu, A.Z., Swenson, G.R. [2003. A modeling study of O2 and OH airglow perturbations induced by atmospheric gravity waves. J. Geophys. Res. 108 (No. D4), 4151. doi:10.1029/2002JD002474]. 相似文献