Comparison of machine learning methods for copper ore grade estimation

In this study, machine learning methods such as neural networks, random forests, and Gaussian processes are applied to the estimation of copper grade in a mineral deposit. The performance of these methods is compared to geostatistical techniques, such as ordinary kriging and indicator kriging. To ensure that these comparisons are realistic and relevant, the predictive accuracy is estimated on test instances located in drill holes that are different from the training data. The results of an extensive empirical study in the Sarcheshmeh porphyry copper deposit in Southeastern Iran illustrate that specially designed Gaussian processes with a symmetric standardization of the spatial location inputs and an anisotropic kernel yield the most accurate predictions. Furthermore, significant improvements are obtained when, besides location, information on the rock type is included in the set of predictor variables. This observation highlights the importance of carrying out detailed studies of the geological composition of the deposit to obtain more accurate ore grade predictions.     

CAD-DRASTIC: chloride application density combined with DRASTIC for assessing groundwater vulnerability to road salt application

Road salt is pervasively used throughout Canada and in other cold regions during winter. For cities relying exclusively on groundwater, it is important to plan and minimize the application of salt accordingly to mitigate the adverse effects of high chloride concentrations in water supply aquifers. The use of geospatial data (road network, land use, Quaternary and bedrock geology, average annual recharge, water-table depth, soil distribution, topography) in the DRASTIC methodology provides an efficient way of distinguishing salt-vulnerable areas associated with groundwater supply wells, to aid in the implementation of appropriate management practices for road salt application in urban areas. This research presents a GIS-based methodology to accomplish a vulnerability analysis for 12 municipal water supply wells within the City of Guelph, Ontario, Canada. The chloride application density (CAD) value at each supply well is calculated and related to the measured groundwater chloride concentrations and further combined with soil media and aquifer vadose- and saturated-zone properties used in DRASTIC. This combined approach, CAD-DRASTIC, is more accurate than existing groundwater vulnerability mapping methods and can be used by municipalities and other water managers to further improve groundwater protection related to road salt application.     

Nonlinear transformation of unit hydrograph

Unit hydrograph (UH) and its numerous derivatives have been popular for estimation of flood hydrographs. Two major assumptions still overshadow UH applications. One is the linearity and the other is time invariance. In theory, only peak discharge of an equilibrium hydrograph follows linear proportionality to excess rainfall intensity. In trying to relax the linearity constraint, this paper aims to propose a nonlinear way of transforming a given UH to other general hydrographs. The transformation or mapping technique relies on a simple rainfall ratio raised to a power less than unity. The case of nonlinear transformation is illustrated for a number of watershed geometries with either known kinematic wave analytic solutions or observed data. The nonlinear UH approach also relaxes the assumption of constant time base of the UH. The proposed nonlinear UH transformation may thus be viewed as a major step in closing the gap between physically based and traditional UH-based surface runoff simulation approaches.     

Near-infrared luminosity function and colours of dwarf galaxies in the Coma cluster

We present K -band observations of the low-luminosity galaxies in the Coma cluster, which are responsible for the steep upturn in the optical luminosity function at M _R∼−16, discovered recently. The main results of this study are as follows.
(i) The optical–near-infrared colours of these galaxies imply that they are dwarf spheroidal galaxies. The median B − K colour for galaxies with −19.3< M_K <−16.3 is 3.6 mag.
(ii) The K -band luminosity function in the Coma cluster is not well constrained, because of the uncertainties due to the field-to-field variance of the background. However, within the estimated large errors, this is consistent with the R -band luminosity function, shifted by ∼3 mag.
(iii) Many of the cluster dwarfs lie in a region of the B − K versus B − R colour–colour diagram where background galaxies are rare ( B − K <5; 1.2< B − R <1.6). Local dwarf spheroidal galaxies lie in this region too. This suggests that a better measurement of the K -band cluster luminosity can be made if the field-to-field variance of the background can be measured as a function of colour, even if it is large.
(iv) If we assume that none of the galaxies in the region of the B − K versus B − R plane given in (iii) in our cluster fields are background, and that all the cluster galaxies with 15.5< K <18.5 lie in this region of the plane, then we measure α=−1.41^+0.34_−0.37 for −19.3< M_K −16.3, where α is the logarithmic slope of the luminosity function. The uncertainties in this number come from counting statistics.     

Backcasting long-term climate data: evaluation of hypothesis

Most often than not, incomplete datasets or short-term recorded data in vast regions impedes reliable climate and water studies. Various methods, such as simple correlation with stations having long-term time series, are practiced to infill or extend the period of observation at stations with missing or short-term data. In the current paper and for the first time, the hypothesis on the feasibility of extending the downscaling concept to backcast local observation records using large-scale atmospheric predictors is examined. Backcasting is coined here to contrast forecasting/projection; the former is implied to reconstruct in the past, while the latter represents projection in the future. To assess our hypotheses, daily and monthly statistical downscaling models were employed to reconstruct past precipitation data and lengthen the data period. Urmia and Tabriz synoptic stations, located in northwestern Iran, constituted two case study stations. SDSM and data-mining downscaling model (DMDM) daily as well as the group method of data handling (GMDH) and model tree (Mp5) monthly downscaling models were trained with National Center for Environmental Prediction (NCEP) data. After training, reconstructed precipitation data of the past was validated against observed data. Then, the data was fully extended to the 1948 to 2009 period corresponding to available NCEP data period. The results showed that DMDM performed superior in generation of monthly average precipitation compared with the SDSM, Mp5, and GMDH models, although none of the models could preserve the monthly variance. This overall confirms practical value of the proposed approach in extension of the past historic data, particularly for long-term climatological and water budget studies.     

Regional analysis of streamflow drought: a case study in southwestern Iran

Droughts are complex natural hazards that, to a varying degree, affect some parts of the world every year. The range of drought impacts is related to drought occurring in different stages of the hydrological cycle and usually different types of droughts such as meteorological, agricultural, hydrological, and socio-economical are the most distinguished types. Hydrological drought includes streamflow and groundwater droughts. In this paper, streamflow drought was analyzed using the method of truncation level (at 70 % level) by daily discharges at 54 stations in southwestern Iran. Frequency analysis was carried out for annual maximum series of drought deficit volume and duration. 35 factors such as physiographic, climatic, geologic and vegetation were studied to carry out the regional analysis. According to conclusions of factor analysis, the six most effective factors include watershed area, the sum rain from December to February, the percentage of area with NDVI <0.1, the percentage of convex area, drainage density and the minimum of watershed elevation, explained 89.2 % of variance. The homogenous regions were determined by cluster analysis and discriminate function analysis. The suitable multivariate regression models were ascertained and evaluated for hydrological drought deficit volume with 2 years return period. The significance level of models was 0.01. The conclusion showed that the watershed area is the most effective factor that has a high correlation with drought deficit volume. Moreover, drought duration was not a suitable index for regional analysis.     

Evaluating the impacts of watershed management on runoff storage and peak flow in Gav-Darreh watershed,Kurdistan, Iran

Recently, water and soil resource competition and environmental degradation due to inadequate management practices have been increased and pose difficult problems for resource managers. Numerous watershed practices currently being implemented for runoff storage and flood control purposes have improved hydrologic conditions in watersheds and enhanced the establishment of riparian vegetation. The assessment of proposed management options increases management efficiency. The purpose of this study is to assess the impact of watershed managements on runoff storage and peak flow, and determine the land use and cover dynamics that it has induced in Gav-Darreh watershed, Kurdistan, Iran. The watershed area is 6.27 km² which has been subjected to non-structural and structural measures. The implemented management practices and its impact on land use and cover were assessed by integrating field observation and geographic information systems (GIS). The data were used to derive the volume of retained water and determine reduction in peak flow. The hydrology of the watershed was modeled using the Hydrologic Engineering Center–Hydrologic Modeling System (HEC–HMS) model, and watershed changes were quantified through field work. Actual storms were used to calibrate and validate HEC–HMS rainfall–runoff model. The calibrated HEC–HMS model was used to simulate pre- and post-management conditions in the watershed. The results derived from field observation and HEC–HMS model showed that the practices had significant impacts on the runoff storage and peak flow reduction.     

Evaluation of dynamic regression and artificial neural networks models for real-time hydrological drought forecasting

In this study, application of a class of stochastic dynamic models and a class of artificial intelligence model is reported for the forecasting of real-time hydrological droughts in the Black River basin in the USA. For this purpose, the Standardized Hydrological Drought Index (SHDI) was adopted in different time scales to represent the hydrological drought index. Six probability distribution functions (PDF) were fitted to the discharge time series to obtain the best fit for SHDI calculation. Then, a dynamic linear spatio-temporal model (DLSTM) and artificial neural network (ANN) were used to forecast SHDI. Although results indicated that both models were able to forecast SHDI in different time scales, the DLSTM was far superior in longer lead times. The DLSTM could forecast SHDI up to 6 months ahead while ANN was only capable of forecasting SHDI up to 2 months ahead appropriately. For short lead times (1–6 months), the DLSTM has performed nearly perfect in test phase and CE oscillates between 0.97 and 0.86 while for ANN modeling, CE is between 0.72 and 0.07. However, the performance of DLSTM and ANN reduced considerably in medium lead times (7–12 months). Overall, the DLSTM is a powerful tool for appropriately forecasting SHDI at short time scales; a major advantage required for drought early warning systems.     

Evaluation of rainfall spatial correlation effect on rainfall-runoff modeling uncertainty,considering 2-copula

Lack of accuracy of rainfall-runoff simulation (RRS) remains critical for some applications. Among various sources of uncertainty, precipitation plays a particular role. Rainfall rates as the main input data of RRS are of the first factors controlling the accuracy. In addition to the depth, spatial and temporal distributions of rainfall impact the flood discharge. Most of the previous studies on RRS uncertainty have ignored rainfall spatial distribution, where in large catchments, it is necessary to be modeled explicitly. Karoon III is one most important basin of the Iran because of the Karoon III dam in the outlet. In the present work, effect of spatial correlation of rainfall on HEC-HMS (SMA) continuous RRS uncertainty is evaluated using 2variate copula (2copula). Monte Carlo simulation (MCS) approach was used to consider the rainfall spatial dependence. To reduce the computational expense, sampling efficiency and convergence for MCS, Latin hypercube sampling (LHS) was used. Copula functions consider wide range of marginal probability distribution functions (PDFs), eliminating limits of regular join PDFs. For this aim, two scenarios were investigated. In the first scenario, sub-basin rainfall was considered independent, and in the second scenario, 2copula was adopted to model spatial correlation of rainfall. Dimensionless rainfall depths were calculated for each sub-basin, and the PDFs were determined. The generated random dimensionless rainfalls were reweighted and multiplied by watershed’s mean rainfall value. Stochastic Climate Library was used to generate continuous daily rainfalls. Sampling from dimensionless rainfalls using LHS algorithm, 100 runs of calibrated model-simulated 100 flows for each day following MCS, and 80 % certainty bound was calculated. Results showed that considering dependence decreased 18 % of the maximum uncertainty bound width, so the methodology could be recommended for decreasing predicted runoff error.