ABSTRACTThe spatio-temporal residual network (ST-ResNet) leverages the power of deep learning (DL) for predicting the volume of citywide spatio-temporal flows. However, this model, neglects the dynamic dependency of the input flows in the temporal dimension, which affects what spatio-temporal features may be captured in the result. This study introduces a long short-term memory (LSTM) neural network into the ST-ResNet to form a hybrid integrated-DL model to predict the volumes of citywide spatio-temporal flows (called HIDLST). The new model can dynamically learn the temporal dependency among flows via the feedback connection in the LSTM to improve accurate captures of spatio-temporal features in the flows. We test the HIDLST model by predicting the volumes of citywide taxi flows in Beijing, China. We tune the hyperparameters of the HIDLST model to optimize the prediction accuracy. A comparative study shows that the proposed model consistently outperforms ST-ResNet and several other typical DL-based models on prediction accuracy. Furthermore, we discuss the distribution of prediction errors and the contributions of the different spatio-temporal patterns. 相似文献
Mineral potential prediction is a process of establishing a statistical model that describes the relationship between evidence variables and mineral occurrences. In this study, evidence variables were constructed from geological, remote sensing, and geochemical data collected from the Lalingzaohuo district, Qinghai Province, China. Based on these evidence variables, a conjugate gradient logistic regression (CG-LR) model was established to predict exploration targets in the study area. The receiver operating characteristic (ROC) and prediction–area (P-A) curves were used to evaluate the effectiveness of the CG-LR model in mineral potential mapping. The difference between the vertical and horizontal coordinates of each point on the ROC curve was used to determine the optimal threshold for classifying the exploration targets. The optimal threshold corresponds to the point on the ROC curve where the difference between the vertical coordinate and the horizontal coordinate is the largest. In exploration target prediction in the study area, the CG algorithm was used to optimize iteratively the LR coefficients, and the prediction effectiveness was tested for different epochs. With increasing iterations, the prediction performance of the model becomes increasingly better. After 60 iterations, the LR model becomes stable and has the best performance in exploration target prediction. At this point, the exploration targets predicted by the CG-LR model occupy 14.39% of the study area and contain 93% of the known mineral deposits. The exploration targets predicted by the model are consistent with the metallogenic geological characteristics of the study area. Therefore, the CG-LR model can effectively integrate geological, remote sensing, and geochemical data for the study area to predict targets for mineral exploration.
Exploring the spatial relationships between various geological features and mineralization is not only conducive to understanding the genesis of ore deposits but can also help to guide mineral exploration by providing predictive mineral maps. However, most current methods assume spatially constant determinants of mineralization and therefore have limited applicability to detecting possible spatially non-stationary relationships between the geological features and the mineralization. In this paper, the spatial variation between the distribution of mineralization and its determining factors is described for a case study in the Dingjiashan Pb–Zn deposit, China. A local regression modeling technique, geological weighted regression (GWR), was leveraged to study the spatial non-stationarity in the 3D geological space. First, ordinary least-squares (OLS) regression was applied, the redundancy and significance of the controlling factors were tested, and the spatial dependency in Zn and Pb ore grade measurements was confirmed. Second, GWR models with different kernel functions in 3D space were applied, and their results were compared to the OLS model. The results show a superior performance of GWR compared with OLS and a significant spatial non-stationarity in the determinants of ore grade. Third, a non-stationarity test was performed. The stationarity index and the Monte Carlo stationarity test demonstrate the non-stationarity of all the variables throughout the area. Finally, the influences of the degree of non-stationary of all controlling factors on mineralization are discussed. The existence of significant non-stationarity of mineral ore determinants in 3D space opens up an exciting avenue for research into the prediction of underground ore bodies.
Natural Resources Research - Some industrial activities, such as underground mining, hydraulic fracturing (HF), can cause microearthquakes and even damaging earthquakes. In recent years, with the... 相似文献
Soil CO_2 efflux, the second largest flux in a forest carbon budget, plays an important role in global carbon cycling. Forest logging is expected to have large effects on soil CO_2 efflux and carbon sequestration in forest ecosystems. However, a comprehensive understanding of soil CO_2 efflux dynamics in response to forest logging remains elusive due to large variability in results obtained across individual studies. Here, we used a meta-analysis approach to synthesize the results of 77 individual field studies to determine the impacts of forest logging on soil CO_2 efflux. Our results reveal that forest logging significantly stimulated soil CO_2 efflux of the growing season by 5.02%. However, averaged across all studies, nonsignificant effect was detected following forest logging. The large variation among forest logging impacts was best explained by forest type, logging type, and time since logging. Soil CO_2 efflux in coniferous forests exhibited a significant increase(4.38%) due to forest logging, while mixed and hardwood forests showed no significant change. Logging type also had a significant effect on soil CO_2 efflux, with thinning increasing soil CO_2 efflux by 12.05%, while clear-cutting decreasing soil CO_2 efflux by 8.63%. The time since logging also had variable effects, with higher soil CO_2 efflux for 2 years after logging, and lower for 3-6 years after logging; when exceeded 6 years, soil CO_2 efflux increased. As significantly negative impacts of forest logging were detected on fine root biomass, the general positive effects on soil CO_2 efflux can be explained by the accelerated decomposition of organic matter as a result of elevated soil temperature and organic substrate quality. Our results demonstrate that forest logging had potentially negative effects on carbon sequestration in forest ecosystems. 相似文献
Redox hot spots occurring as metal-rich anoxic groundwater discharges through oxic wetland and river sediments commonly result in the formation of iron (Fe) oxide precipitates. These redox-sensitive precipitates influence the release of nutrients and metals to surface water and can act as ‘contaminant sponges’ by absorbing toxic compounds. We explore the feasibility of a non-invasive, high-resolution magnetic susceptibility (MS) technique to efficiently map the spatial variations of magnetic Fe oxide precipitates in the shallow bed of three rivers impacted by anoxic groundwater discharge. Laboratory analyses on Mashpee River (MA, USA) sediments demonstrate the sensitivity of MS to sediment Fe concentrations. Field surveys in the Mashpee and Quashnet rivers (MA, USA) reveal several discrete high MS zones, which are associated with likely anoxic groundwater discharge as evaluated by riverbed temperature, vertical head gradient, and groundwater chemistry measurements. In the East River (CO, USA), widespread cobbles/rocks exhibit high background MS from geological ferrimagnetic minerals, thereby obscuring the relatively small enhancement of MS from groundwater induced Fe oxide precipitates. Our study suggests that, in settings with low geological sources of magnetic minerals such as lowland rivers and wetlands, MS may serve as a complementary tool to temperature methods for efficiently mapping Fe oxide accumulation zones due to anoxic groundwater discharges that may function as biogeochemical hot spots and water quality control points in gaining systems. 相似文献
ABSTRACTThis study investigated the late Quaternary climate and environmental characteristics of two tributary valleys (Xingmu and Depu Valleys) in the Parlung Zangbo Valley, southeastern Tibetan Plateau. Optically stimulated luminescence (OSL) samples collected from moraines at the mouth of Xingmu Valley produce a wide age range from 13.9 ka to 76 ka. The ages measured from the lenticular sand are consistent with the relative geomorphic sequence of the landforms. Lenticular sand layers below the moraine were dated to 37.9 ka and 44.7 ka, indicating that fluvial processes were likely dominant in the valley during Marine Isotope Stage (MIS) 3. The outer moraine ridges at the valley mouth were formed during 13.9 ka and 26.5 ka, corresponding to MIS2. At Depu Valley, OSL samples from two sets of lateral and terminal moraines close to the modern glacier, provide ages from 1.4 ka to 29.2 ka. The paleosol layer widely developed during 2.6 cal ka BP and 8.7 cal ka BP in the study area, reflecting a relatively warm condition during the mid-Holocene. 相似文献
Classic porphyry Cu–Mo deposits are mostly characterized by close temporal and spatial relationships between Cu and Mo mineralization. The northern Dabate Cu–Mo deposit is a newly discovered porphyry Cu–Mo polymetallic deposit in western Tianshan, northwest China. The Cu mineralization postdates the Mo mineralization and is located in shallower levels in the deposit, which is different from most classic porphyry Cu–Mo deposits. Detailed field investigations, together with microthermometry, laser Raman spectroscopy, and O‐isotope studies of fluid inclusions, were conducted to investigate the origin and evolution of ore‐forming fluids from the main Mo to main Cu stage of mineralization in the deposit. The results show that the ore‐forming fluids of the main Mo stage belonged to an NaCl + H2O system of medium to high temperatures (280–310°C) and low salinities (2–4 wt% NaCl equivalent (eq.)), whereas that of the main Cu stage belonged to an F‐rich NaCl + CO2 + H2O system of medium to high temperatures (230–260°C) and medium to low salinities (4–10 wt% NaCl eq.). The δ18O values of the ore‐forming fluids decrease from 3.7–7.8‰ in the main Mo stage to ?7.5 to ?2.9‰ in the main Cu stage. These data indicate that the separation of Cu and Mo was closely related to a large‐scale vapor–brine separation of the early ore‐forming fluids, which produced the Mo‐bearing and Cu‐bearing fluids. Subsequently, the relatively reducing (CH4‐rich) Mo‐bearing, ore‐forming fluids, dominantly of magmatic origin, caused mineralization in the rhyolite porphyry due to fluid boiling, whereas the relatively oxidizing (CO2‐rich) Cu‐bearing, ore‐forming fluids mixed with meteoric water and precipitated chalcopyrite within the crushed zone at the contact between rhyolite porphyry and wall rock. We suggest that the separation of Cu and Mo in the deposit may be attributed to differences in the chemical properties of Cu and Mo, large‐scale vapor–brine separation of early ore‐forming fluids, and changes in oxygen fugacity. 相似文献
The West Junggar Orogenic Belt(WJOB)in northwestern Xinjiang,China,is located in the core of the western part of the Central Asian Orogenic Belt(CAOB).It has suffered two stage tectonic evolutions in Phanerozoic,before and after the ocean–continental conversion in Late Paleozoic.The later on intracontinental deformation,characterized by the development of the NE-trending West Junggar sinistral strike-slip fault system(WJFS)since Late Carboniferous and Early Permian,and the NW-trending Chingiz-Junggar dextral strike-slip fault(CJF)in Mesozoic and Cenozoic,has an important significance for the tectonic evolution of the WJOB and the CAOB.In this paper,we conduct geometric and kinematic analyses of the WJOB,based on field geological survey and structural interpretation of remote sensing image data.Using some piercing points such as truncated plutons and anticlines,an average magnitude of^73 km for the left-lateral strike-slip is calculated for the Darabut Fault,a major fault of the WJFS.Some partial of the displacement should be accommodated by strike-slip fault-related folds developed during the strike-slip faulting.Circular and curved faults,asymmetrical folds,and irregular contribution of ultramafic bodies,implies potential opposite vertical rotation of the Miao’ergou and the Akebasitao batholiths,resulted from the sinistral strike-slipping along the Darabut Fault.Due to conjugate shearing set of the sinistral WJFS and the dextral CJF since Early Mesozoic,superimposed folds formed with N–S convergence in southwestern part of the WJOB. 相似文献