Development of a convolutional neural network based geomechanical upscaling technique for heterogeneous geological reservoir

Geomechanical assessment using coupled reservoir-geomechanical simulation is becoming increasingly important for analyzing the potential geomechanical risks in subsurface geological developments.However,a robust and efficient geomechanical upscaling technique for heterogeneous geological reservoirs is lacking to advance the applications of three-dimensional(3D)reservoir-scale geomechanical simulation considering detailed geological heterogeneities.Here,we develop convolutional neural network(CNN)proxies that reproduce the anisotropic nonlinear geomechanical response caused by lithological heterogeneity,and compute upscaled geomechanical properties from CNN proxies.The CNN proxies are trained using a large dataset of randomly generated spatially correlated sand-shale realizations as inputs and simulation results of their macroscopic geomechanical response as outputs.The trained CNN models can provide the upscaled shear strength(R^(2)>0.949),stress-strain behavior(R^(2)>0.925),and volumetric strain changes(R^(2)>0.958)that highly agree with the numerical simulation results while saving over two orders of magnitude of computational time.This is a major advantage in computing the upscaled geomechanical properties directly from geological realizations without the need to perform local numerical simulations to obtain the geomechanical response.The proposed CNN proxybased upscaling technique has the ability to(1)bridge the gap between the fine-scale geocellular models considering geological uncertainties and computationally efficient geomechanical models used to assess the geomechanical risks of large-scale subsurface development,and(2)improve the efficiency of numerical upscaling techniques that rely on local numerical simulations,leading to significantly increased computational time for uncertainty quantification using numerous geological realizations.

Script-Based GPU-Ready ELM Development for Continuous Code Integration

Designing and optimizing complex scientific code for new computing architectures is a challenging task. To address this issue in the E3SM land model (ELM) development, we developed a software tool called SPEL, which facilitates code generation, verification, and performance tuning using compiler directives within a Function Unit Test framework. In this paper, we present a SPEL extension that leverages the version control system (e.g., Git) to autonomous code generation and demonstrate its application to continuous code integration and development of the ELM software system. The study can benefit the scientific software development community.

Biological transformation,kinetics and dose-response assessments of bound musk ketone hemoglobin adducts in rainbow trout as biomarkers of environmental exposure

Low levels (ng/g) of musk ketone (MK),used as a fragrance additive in the formulation of personal care products,are frequently detected in the water and other environment.Thus,aquatic organisms can be continuously exposed to MK.In this study,kinetics and dose-response assessments of 2-amino-MK (AMK) metabolite,bound to cysteine-hemoglobin (Hb) in rainbow trout,formed by enzymatic nitro-reduction of MK have been demonstrated.Trout were exposed to a single exposure of 0.010,0.030,0.10,and 0.30 mg MK/g fish.Tw...

Hybrid energy module for remote environmental observations, experiments, and communications

Increased concerns about climate change have led to a significant expansion of monitoring, observational, and experimental sites in remote areas of the world. Meanwhile, advances in technology and availability of low-power equipment have allowed increasingly sophisticated measurements with a wide variety of instruments. However, the deployment and use of these technologies in remote locations is often restricted not only by harsh environmental conditions, but also by the availability of electrical power and communication options. In some cases, research stations and military installations can provide power for scientific equipment, data acquisition, storage, and transmission. Clustering of research sites near existing infrastructure has had the unintended consequence of limiting a spatial understanding of large geographic regions. Fortunately, the modern market offers many power and communication solutions, but most of them are oriented toward large industrial applications. Use of those solutions to power a research site is limited because of their cost and need for significant modification for the specific research purposes. Each study has its own unique power requirements and needs for proper instrumentation. A power and communication solution for a vast majority of implementations with or without modification would be of considerable benefit. This article describes design of a universal, scalable hybrid energy module for the Next-Generation Ecosystem Experiments Arctic project(https://ngee-arctic.ornl.gov/). Two modules were built, and the authors describe their implementation and findings over a 2-year period at a remote field site on the Seward Peninsula in western Alaska, USA.

Technical, Economic and Environmental Evaluation on Mechanical Rice Straw Gathering Method

Rice straw is a rice by-product, which is currently mostly wasted in Vietnam, in particular in the Mekong delta. At present, the cost of straw gathering is increasing because of the increased use of combine harvesters. High labor cost and lack of labor makes manual collection unfeasible. Farmers therefore often just burn it, which causes pollution, increased greenhouse gas emissions and loss of opportunities to value add. An economic and environmental evaluation and technical field testing of a straw baler with 4 ha/day capacity was therefore conducted in Long An province. During the field testing data on the gathering capacity, fuel consumption, labor requirement and other cost items were collected. The test results showed that the baling cost is US$19.0 per ton of rice straw, the pay-back period of 2. 1 years and the internal rate of return of 38%. In addition to the baling cost, the transportation cost varies from US$24 for a distance of 100 km to US$32 for 150 km. The benefits of the machine are not only economical but also include the reduction of field burning.

超级杂交稻干物质生产特点与产量稳定性研究

【目的】探明超级杂交稻在不同种植地点和不同施肥量条件下的产量表现及干物质生产特点。【方法】于2004～2005年在湖南省桂东、长沙、衡阳、南县和永州5个地点进行大田试验,按照N﹕P2O5﹕K2O为1﹕0.5﹕1的比例,设置3种施肥量处理(135、180、225kgN·ha-1),田间采用随机区组排列,4次重复,以超级杂交稻组合准两优527和两优293为试验材料。【结果】超级杂交稻收获产量以桂东点产量最高,地点间差异显著,其中准两优527平均为7492.3～12209.2kg·ha-1,两优293为6984.0～11679.5kg·ha-1。产量构成因子和干物质生产量的地点间变化与收获产量一致,但在同一地点的不同施肥量处理间收获产量和干物质生产量差异均不显著。收获产量与单位面积穗数、结实率和千粒重表现为正相关,而与每穗粒数表现为负相关。【结论】超级杂交稻存在适宜的种植区域,且在施肥量为135～225kgN·ha-1的范围内,施肥量不是超高产栽培的限制因子。超级杂交稻的库容量大,提高结实率和粒重是获得超高产的可能途径。

应用集成BP神经网络进行田间土壤空间变异研究

以英国北爱尔兰Hayes的一块牧草地为研究区,将所有样点分为独立的训练和检验数据集,并在训练样点集的基础上设计了其他4种样点布局方案,以研究神经网络集成技术应用于田间土壤性质空间变异性的可能性。与广泛应用的克里格法的试验结果相比,集成BP神经网络的插值结果精度与之基本相当,尤其是在样点分布较稀疏和样点数较少的情况下,集成BP网络表现出明显的优势;由于神经网络集成方法对样本数据的分布没有任何要求,因此具有较广泛的应用前景和潜力,并在不符合克里格法对样本数据分布要求的情况下是一种可行的替代方法。

不同林型土壤微生物有机碳降解基因的多样性

应用寡聚核苷酸基因芯片,分析了米亚罗林区冷杉原始林(M-Y)和20世纪60年代云杉人工林(M-60)土壤微生物的功能基因多样性。该功能基因芯片含有与有机碳降解、碳固定、氮、磷、硫循环和金属抗性相关的1961个基因探针。在M-Y和M-60样地中分别检测到39和62个具有较强杂交信号(SNR≥2)的功能基因,其基因多样性水平指数分别为3.59和4.04,杂交信号强度总值分别为480280和630560。M-Y和M-60样地中分别检测到32个和37个有机碳降解基因,占总基因的82%和60%,这些基因分属于22个不同的基因类群,分别参与木质素、木聚糖、几丁质等有机碳的降解过程。有机碳降解基因在两个样地中存在较大的多样性和丰度差异。这些结果说明了大多数的土壤微生物直接参与了土壤有机碳的降解,同时,林型不同显著影响了土壤微生物群落结构和有机碳降解微生物的多样性。

美国理论生态学研究动态

当今,由于人们对资源、环境等世界性问题的日益关注,因而生态学的发展越来越受到重视。但是,在生态学繁荣发展的背后,却隐藏着一个危机,即对生态学理论的研究还很不够。现在在美国,理论生态学的研究正逐渐地受到重视,其中著名的对生态学的发展起过重要作用,并首先倡导生态数学建模及系统分析研究工作的美国能源部。

Predicting Nitrogen Status of Rice Using Multispectral Data at Canopy Scale

Two field experiments were conducted in Jiashan and Yuhang towns of Zhejiang Province, China, to study the feasibility of predicting N status of rice using canopy spectral reflectance. The canopy spectral reflectance of rice grown with different levels of N inputs was determined at several important growth stages. Statistical analyses showed that as a result of the different levels of N supply, there were significant differences in the N concentrations of canopy leaves at different growth stages. Since spectral reflectance measurements showed that the N status of rice was related to reflectance in the visible and NIR (near-infrared) ranges, observations for rice in 1 nm bandwidths were then converted to bandwidths in the visible and NIR spectral regions with IKONOS (space imaging) bandwidths and vegetation indices being used to predict the N status of rice. The results indicated that canopy reflectance measurements converted to ratio vegetation index (RVI) and normalized difference vegetation index (NDVI) for simulated IKONOS bands provided a better prediction of rice N status than the reflectance measurements in the simulated IKONOS bands themselves. The precision of the developed regression models using RVI and NDVI proved to be very high with R2 ranging from 0.82 to 0.94, and when validated with experimental data from a different site, the results were satisfactory with R2 ranging from 0.55 to 0.70. Thus, the results showed that theoretically it should be possible to monitor N status using remotely sensed data.

Effect of irrigation regime on grain yield,water productivity,and methane emissions in dry direct-seeded rice grown in raised beds with wheat straw incorporation

Dry direct-seeded rice grown in raised beds is becoming an important practice in the wheat–rice rotation system in China.However,little information has been available on the effect of various irrigation regimes on grain yield,water productivity(WP),nitrogen use efficiency(NUE),and greenhouse gas emission in this practice.This study investigated the question using two rice cultivars in 2015 and 2016 grown in soil with wheat straw incorporated into it.Rice seeds were directly seeded into raised beds,which were maintained under aerobic conditions during the early seedling period.Three irrigation regimes:continuous flooding(CF),alternate wetting and drying(AWD),and furrow irrigation(FI),were applied from 4.5-leaf-stage to maturity.Compared with CF,both AWD and FI significantly increased grain yield,WP,and internal NUE,with greater increases under the FI regime.The two cultivars showed the same tendency in both years.Both AWD and FI markedly increased soil redox potential,root and shoot biomass,root oxidation activity,leaf photosynthetic NUE,and harvest index and markedly decreased global warming potential,owing to substantial reduction in seasonalThe results demonstrate that adoption of either AWD or FI could increase grain yield and resource-use efficiency and reduce environmental risks in dry direct-seeded rice grown on raised beds with wheat straw incorporation in the wheat–rice rotation system.

Improvements of a Dynamic Global Vegetation Model and Simulations of Carbon and Water at an Upland-Oak Forest

The interest in the development and improvement of dynamic global vegetation models （DGVMs）, which have the potential to simulate fluxes of carbon, water and nitrogen, along with changes in the vegetation dynamics, within an integrated system, has been increasing. In this paper, some numerical schemes and a higher resolution soil texture dataset were employed to improve the Sheffield Dynamic Global Vegetation Model （SDGVM）. Using eddy covariance-based measurements, we then tested the standard version of the SDGVM and the modified version of the SDGVM. Detailed observations of daily carbon and water fluxes made at the upland oak forest on the Walker Branch Watershed in Tennessee, USA offered a unique opportunity for these comparisons. The results revealed that the modified version of the SDGVM did a reasonable job of simulating the carbon and water flux and the variation of soil water content （SWC）. However, at the end of the growing season, it failed to simulate the effect of the limitations on the soil respiration dynamics and as a result underestimated this respiration. It was also noted that the modified version overestimated the increase in the SWC following summer rainfall, which was attributed to an inadequate representation of the ground water and thermal cycle.

Eco-morphological characteristics of fern species for slope conservation

Degradation of slopes due to shallow landslide and the subsequent erosional processes are a big challenge on the application of soil bioengineering techniques; that is the use of plants as main structural components of a slope protection and conservation system. An optimal application of soil bioengineering techniques should include not only the technical factor of plants as structural components but also the ecology of species and the plant adaptations to disturbances, which is crucial if a longterm successful slope restoration system is intended. Ferns are a dominant understory vegetation species in the forest of Japan, but its characteristics and influences on the recovery of shallow landslide scars have not been fully studied yet. This study aims to find out the ecological characteristics of fern species through the calculation of ecological indicators and the quantification of the morphological features of specimens growing on disturbed and non-disturbed forest slopes in Japan. Gleichenia japonica was found as the vegetation species with biggest ecological indicators on both slopes. The analysis of morphological characteristics of the specimens growing on both sites showed that the development of the specimens is focused in below-ground characteristics. The pull-out force of Gleichenia japonica root system as an indicator of ecological adaptation to a constraint environment and morphological characteristics quality is influenced by height and root length according to the principal component analysis. The eco-morphological characteristics of species can be used as an indicator of an optimal element in soil bioengineering establishment for slope conservation proposes. The long and fibrous root system could be placed on forest roads, steep or small slopes where space limitation is an issue for the establishment of bigger species and if the slope conditions allow it, it can control soil losses due to rainfall and provide stability.

Caffeine Crystallization Induction Time Measurements Using Laser Scattering Technique and Correlation to Surface Tension in Water and Ethanol

Caffeine nucleation induction times were measured at 30 °C and 40 °C in water and ethanol solvents employing laser light absorption technique. Supersaturation concentrations and liquid/solid phase surface tensions were calculated from crystallization induction times using classic homogeneous nucleation theory. Induction time and surface tension decreased at higher temperature.

Morpho-Physiological Changes in Roots of Rice Seedling upon Submergence

Submergence is a serious environmental condition that causes large loss in rice production in rain fed lowland and flood affected area. This study evaluated morphological and physiological responses of rice roots to submergence using two tolerant rice genotypes FR13A and Swarna-Sub 1 and two sensitive ones Swarna and IR42. The tolerant genotypes had higher survival rate and less shoot elongation but greater root elongation during submergence than the sensitive ones. After submergence,the tolerant genotypes also had higher root dry weight and more active roots than the sensitive ones.Tolerant genotypes exhibited less root injury, with less malondialdehyde production and slower electrolyte leakage after submergence. Tolerant genotypes also maintained higher concentrations of soluble sugar and starch in roots and shoots and higher chlorophyll retention after submergence than the sensitive ones.Our data showed that root traits such as root activity and root growth are associated with survival rate after submergence. This is probably accomplished through higher energy supply, and membrane integrity is necessary to preserve root function and reduce injury during submergence. These root traits are important for submergence tolerance in rice.

Responses of Contrasting Rice (Oryza sativa L.) Genotypes to Salt Stress as Affected by Nutrient Concentrations

The study was conducted to investigate the effects of applying different concentrations of the macronutrients K＋,Ca2＋,and Mg2＋ on the responses of contrasting rice（Oryza sativa L.） genotypes under salt stress.A solution culture experiment was conducted in a phytotron at the International Rice Research Institute（IRRI）,under controlled temperature and humidity and natural sunlight.When subjected to salt stress of 100 mmol L-1 using NaCl,the salt tolerant genotypes FL478 and IR651,accumulated less Na＋ and maintained lower ratios of Na＋/K＋,Na＋/Ca2＋,and Na＋/Mg2＋ than the sensitive genotypes IR29 and Azucena.These tolerant genotypes also had higher concentrations of K＋ in their shoots and greater root and shoot biomass and green leaf area.Tolerant genotypes also maintained much lower concentration of Na＋ and lower and more favorable ratios of Na＋/K＋,Na＋/Ca2＋,and Na＋/Mg2＋ in their active and developing tissues.Salt tolerance and shoot and root growth of both tolerant and sensitive genotypes were enhanced considerably when higher concentrations of Ca2＋ and Mg2＋ were applied in culture solution.The concentration of Na＋ and the ratios of Na＋/K＋,Na＋/Ca2＋,and Na＋/Mg2＋ in shoots also declined significantly.The beneficial effects of higher calcium were greater than that of magnesium and application of higher concentration of K＋ seems to have minor effects.Responses to salinity in rice can therefore be considerably enhanced through proper nutrient management,by increasing the concentrations of nutrient elements that have favorable effects such as Ca2＋ and Mg2＋.Calcium is particularly more effective than both magnesium and potassium,and can be applied at relatively larger quantities in salt affected soils.

Carbon sequestration of plantation in Beijing-Tianjin sand source areas

The Beijing-Tianjin Sand Source Control Project(BTSSCP), a national ecological restoration project, was launched to construct an ecological protection system in the Beijing-Tianjin sand source areas to reduce dust hazards. The carbon sequestration dynamics can be used to assess the ecological effects of an ecological restoration project. Here, we conducted vegetation and soil study to assess the carbon sequestration in the plantations with 10 years old stands in Beijing-Tianjin sand source areas. The results at the site scales indicated that the average net increase of plantation ecosystem carbon stock was 33.8 Mg C ha^(-1), with an annual increase rate of 3.38 Mg C ha^(-1) yr^(-1). The average net increase of carbon varied among regions, vegetation types, and forest management activities. Soil bulk density in the top soil decreased slightly after 10-year implementation of the project. Coniferous forests and shrubs are suitable plant species for sand source areas.Natural restoration in the plantations is a practical and feasible and promising approach for enhancing ecosystem carbon sequestration potential.

Active forest management accelerates carbon storage in plantation forests in Lishui,southern China

Background:China has committed to achieving peak CO_(2)emissions before 2030 and carbon neutrality before 2060;therefore,accelerated efforts are needed to better understand carbon accounting in industry and energy fields as well as terrestrial ecosystems.The carbon sink capacity of plantation forests contributes to the mitigation of climate change.Plantation forests throughout the world are intensively managed,and there is an urgent need to evaluate the effects of such management on long-term carbon dynamics.Methods:We assessed the carbon cycling patterns of ecosystems characterized by three typical plantation species(Chinese fir(Cunninghamia lanceolata(Lamb.)Hook.),oak(Cyclobalanopsis glauca(Thunb.)Oerst.),and pine(Pinus massoniana Lamb.))in Lishui,southern China,by using an integrated biosphere simulator(IBIS)tuned with localized parameters.Then,we used the state-and-transition simulation model(STSM)to study the effects of active forest management(AFM)on carbon storage by combining forest disturbance history and carbon cycle regimes.Results:1)The carbon stock of the oak plantation was lower at an early age(<50 years)but higher at an advanced age(>50 years)than that of the Chinese fir and pine plantations.2)The carbon densities of the pine and Chinese fir plantations peaked at 70 years(223.36 Mg⋅ha^(‒1))and 64 years(232.04 Mg⋅ha^(‒1)),respectively,while the carbon density in the oak plantation continued increasing(>100 years).3)From 1989 to 2019,the total carbon pools of the three plantation ecosystems followed an upward trend(an annual increase of 0.16–0.22 Tg C),with the largest proportional increase in the aboveground biomass carbon pool.4)AFM increased the recovery of carbon storage after 1996 and 2009 in the pine and Chinese fir plantations,respectively,but did not result in higher growth in the oak plantation.5)The proposed harvest planning is reasonable and conducive to maximizing the carbon sequestration capacity of the forest.Conclusions:This study provides an example of a carbon cycle coupling model that is potentially suitable for simulating China's plantation forest ecosystems and supporting carbon accounting to monitor peak CO_(2)emissions and reach carbon neutrality.

Waterlogging risk assessment based on self-organizing map(SOM)artificial neural networks:a case study of an urban storm in Beijing

Due to rapid urbanization, waterlogging induced by torrential rainfall has become a global concern and a potential risk affecting urban habitant's safety. Widespread waterlogging disasters haveoccurred almost annuallyinthe urban area of Beijing, the capital of China. Based on a selforganizing map(SOM) artificial neural network(ANN), a graded waterlogging risk assessment was conducted on 56 low-lying points in Beijing, China. Social risk factors, such as Gross domestic product(GDP), population density, and traffic congestion, were utilized as input datasets in this study. The results indicate that SOM-ANNis suitable for automatically and quantitatively assessing risks associated with waterlogging. The greatest advantage of SOM-ANN in the assessment of waterlogging risk is that a priori knowledge about classification categories and assessment indicator weights is not needed. As a result, SOM-ANN can effectively overcome interference from subjective factors,producing classification results that are more objective and accurate. In this paper, the risk level of waterlogging in Beijing was divided into five grades. The points that were assigned risk grades of IV or Vwere located mainly in the districts of Chaoyang, Haidian, Xicheng, and Dongcheng.