Multiple Regression Algorithm Research Articles

Mapping forest height using photon-counting LiDAR data and Landsat 8 OLI data: A case study in Virginia and North Carolina, USA

• Mapping forest height by combining photon-counting LiDAR data and Landsat OLI data. • Douglas-Peucker algorithm can solve the large turns in photon-counting LiDAR data. • The filtering of forest heights can improve the accuracy of forest height map. Large-scale, accurate and detailed forest height map is worthwhile and necessary for understanding and assessing global carbon cycle and biodiversity. The Ice, Cloud and Land Elevation Satellite-2 (ICESat-2) mission, employing a photon-counting LiDAR (PCL) system, offers an opportunity to map global forest height with high resolution. This study aimed to map forest height at a spatial resolution of 30 m by combining Multiple Altimeter Beam Experimental Lidar (MABEL) data with Landsat 8 Operational Land Imager (OLI) data. There are four key steps to accomplish this goal. First, a segmentation method based on the Douglas-Peucker algorithm was proposed to solve the problem of large turns or calibration maneuvers in MABEL data. Second, we estimated forest height and selected the forest height samples with high accuracy and reliability by developing three filters including signal-to-noise ratio (SNR) filter, slope filter, and canopy photons density (CPD) filter. Third, forest height models based on both random forest (RF) and stepwise multiple regression algorithms were developed to establish relationships between the selected forest height samples and predicator variables of Landsat-derived spectral indices, topographic variables and geographic coordinates. Finally, a wall-to-wall forest height map was generated by applying the developed forest height models to predicator variables, and the accuracy of forest height map was validated using airborne LiDAR-derived forest heights. An area of 160, 000 km 2 in southeast Virginia and east North Carolina was chosen for testing the methods proposed in this study. The results demonstrated that the Douglas-Peucker algorithm can effectively solve the MABEL data overlapping issues caused by large turns and calibration maneuvers in flight lines. The suitable filters for selecting forest heights are SNR > 6, terrain slope < 25°, and 40 < CPD < 170. These developed filters can substantially increase the accuracies of forest height models. The results also indicated that RF-derived forest height models achieved higher modeling accuracy than stepwise regression-derived forest height models. RF-derived forest height models yielded coefficient of determination (R 2 ) values of 0.59, 0.68 and 0.62 and RMSE values of 4.55 m, 3.41 m and 4.08 m for deciduous forest, evergreen forest and mixed forest, respectively. Compared to LiDAR-derived forest height, the forest height map produced in this study has a R 2 value of 0.54 and a RMSE value of 6.85 m, which demonstrates that combination of MABEL data and Landsat 8 OLI data can be used to generate forest height maps with a spatial resolution of 30 m.

Practical Applications of Be Wary of Black-Box Trading Algorithms

<h3>Practical Applications Summary</h3> In <b><i>Be Wary of Black-Box Trading Algorithms</i></b>, from the August 2019 issue of <b><i>The Journal of Investing</i></b>, <b>Gary Smith</b> of <b>Pomona College</b> argues that investors should not blindly trust trading algorithms, especially those that are totally opaque. The growing trend of relying on artificial intelligence to conduct trading activities is ill advised, he warns, because computer “intelligence” is very different from human intelligence. Although computers are excellent at detecting statistical relationships, they are terrible at evaluating the merit of those relationships. Smith urges investors to continue to rely on human judgment to evaluate correlations discovered by trading algorithms. Those judgments must be based on sound theory, Smith says, since even the most highly correlated trading models will fail to predict future performance without a solid theoretical basis. He provides an example of a simple model discovered by a multiple regression algorithm that correlated strongly with 2016 data but diverged sharply from 2017 data. This example provides evidence for the dangers of pursuing trading models based purely on data rather than logic or reliable theory. <b>TOPICS:</b>Quantitative methods, risk management

Towards an automated decision support system for the identification of additive manufacturing part candidates

As additive manufacturing (AM) continues to mature, an efficient and effective method to identify parts which are eligible for AM as well as gaining insight on what values it may add to a product is needed. Prior methods are naturally developed and highly experience-dependent, which falls short for its objectiveness and transferability. In this paper, a decision support system (DSS) framework for automatically determining the candidacy of a part or assembly for AM applications is proposed based on machine learning (ML) and carefully selected candidacy criteria. With the goal of supporting efficient candidate screening in the early conceptual design stage, these criteria are further individually decoded to decisive parameters which can be extracted from digital models or resource planning databases. Over 200 existing industrial examples are manually collected and labelled as training data; meanwhile, multiple regression algorithms are tested against each AM potential to find better predictive performance. The proposed DSS framework is implemented as a web application with integrated cloud-based database and ML service, which allows advantages of easy maintenance, upgrade, and retraining of ML models. Two case studies of a hip implant and a throttle pedal are used as demonstrating examples. This preliminary work provides a promising solution for lowering the requirements of non-AM experts to find suitable AM candidates.

QUANTITATIVE INVERSION OF CHLOROPHYLL A CONCENTRATION AND ANALYSIS OF TEMPORAL AND SPATIAL CHARACTERISTICS IN HONGKONG OFFSHORE WATERS BASED ON TIME-SERIES HJ-1A/B MULTISPECTRAL DATA

Abstract. In order to meet the demand of obtaining chlorophyll a concentration distribution in the offshore waters of Hong Kong on a rapid and large scale, based on the daily measured chlorophyll a concentration data and the HJ-1A/B satellite multispectral data in 2009–2019. Taking Modis-Aqua and Himawari series chlorophyll a concentration of remote sensing products as comparative data. Through per-process to extracted the reflectivity of each band combined and statistical correlation analysis with measured data, select the highest correlation band combination as the characteristic variable to statistical regression analysis with two-thirds of the measured chlorophyll a concentration data, constructing multiple regression algorithms. Determination of the best inversion algorithm by verified accuracy of the remaining one-third of the measured data. Inversion of chlorophyll a concentration in the offshore waters of Hong Kong from 2009 to 2019, the results show: 1). The best inversion algorithm is a univariate cubic regression algorithm, R2 is 0.942, RMSE is 0.285 μg/L, RPD is 33%; 2). The overall trend of remote products is the same as the measured data, but the value is slightly higher than the measured data;3). The distribution characteristics in the offshore waters of Hong Kong in the recent ten years are “high in the east and low in the west”, and the average concentration in the western waters is about 5 μg/L lower than the eastern waters.

Paleohydrology and Machine-Assisted Estimation of Paleogeomorphology of Fluvial Channels of the Lower Middle Pennsylvanian Allegheny Formation, Birch River, WV

Rivers transport sediments in a source to sink system while responding to allogenic controls of the depositional system. Stacked fluvial sandstones of the Middle Pennsylvanian (Desmoinesian Stage, ~310 – 306 Ma) Allegheny Formation (MPAF) exposed at Birch River, West Virginia exhibit change in sedimentary structure and depositional style, reflecting changes in allogenic behavior. Paleohydrologic and numerical analysis were used to quantify geomorphological and paleohydrologic variations reflected by MPAF fluvial deposits with the goal of understanding the controls on resulting fluvial sandstone architecture in these different systems. Channel body geometry, sedimentary structures, and sandstone grain size distribution were used to reconstruct the paleoslope and flow velocity of the MPAF fluvial systems. In order to enhance paleohydrological estimates, machine learning methods including multiple regression and Support Vector Regression (SVR) algorithms were used to improve the dune height, and channel depth estimated from cross-set thickness. Results show that the Channel depths of the lower MPAF beneath the Lower Kittanning coal beds tend to decrease upsection; this decrease may reflect a transition from fluvial systems formed in a humid ever-wet climate to fluvial systems formed in less humid, seasonally wet, semi-arid climate. Paleohydraulic estimations enabled the evaluation of hydraulic changes in the fluvial depositional systems of the Appalachian Basin during the Desmoinesian stage. Paleoslope estimates indicated that the slope was low which indicated that the fluvial gradient response was not driven by the effect of tectonic subsidence or uplift and sea-level change.

Deep Learning based Multiple Regression to Predict Total Column Water Vapor (TCWV) from Physical Parameters in West Africa by using Keras Library

Total column water vapor is an important factor for the weather and climate. This study apply deep learning based multiple regression to map the TCWV with elements that can improve spatiotemporal prediction. In this study, we predict the TCWV with the use of ERA5 that is the fifth generation ECMWF atmospheric reanalysis of the global climate. We use an appropriate deep learning based multiple regression algorithm using Keras library to improve nonlinear prediction between Total Column water vapor and predictors as Mean sea level pressure, Surface pressure, Sea surface temperature, 100 metre U wind component, 100 metre V wind component, 10 metre U wind component, 10 metre V wind component, 2 metre dew point temperature, 2 metre temperature. The results obtained permit to build a predictor which modelling TCWV with a mean abs error (MAE) equal to 3.60 kg/m2 and a coefficient of determination R2 equal to 0.90.

Machine learning models for predicting ship main engine Fuel Oil Consumption: A comparative study

As Fuel Oil Consumption (FOC) constitutes over 25% of a vessel’s overall operating cost, its accurate forecasting, and the reliable prediction of the relevant ship operating expenditures can majorly impact the ship operation sustainability and profitability. This study presents a comparison of data-driven, multiple regression algorithms for predicting ship main engine FOC considering two different shipboard data acquisition strategies, noon-reports and Automated Data Logging & Monitoring (ADLM) systems. For this, various multiple regression algorithms including Support Vector Machines (SVMs), Random Forest Regressors (RFRs), Extra Trees Regressors (ETRs), Artificial Neural Networks (ANNs), and ensemble methods are employed. The effectiveness of the tested algorithms is investigated based on a number of key performance indicators, such as the mean and median average error and the coefficient of determination (R2). ETR and RFR models were found to perform best in both cases, whilst the existence of an ADLM system increased accuracy by 7% and reduced the required period for data collection by up to 90%. The derived models can accurately predict the FOC of vessels sailing under different load conditions, weather conditions, speed, sailing distance, and drafts.

Data-driven approach to prediction of residential energy consumption at urban scales in London

Development of energy prediction models plays an integral part in management and enhancement of the energy efficiency of buildings, including carbon emission reduction. Simplified and data-driven models are often the preferred option when detailed information of simulation is not available and the fast responses are required. This study developed data-driven models for predicting electricity and gas consumption in London’s residential buildings at the middle super output areas (MSOA) and lower super output areas (LSOA) with multilayer neural network (MNN), multiple regression (MLR), random forest (RF), and gradient boosting (GB) algorithms, and factors related to socio-demographic, economic, and building characteristics were used as predictors. The results revealed that building characteristics, household income, and the number of households were the most important predictors of electricity and gas consumption. We also found that MNN models have outperformed MLR, RF and GB models in electricity and gas consumption prediction at MSOA and LSOA levels, with R2 values over 0.99 for the electricity consumption model. In summary, this study shows that the MNN models can be a useful tool to assist the formation of energy efficiency policies in buildings at MSOA and LSOA levels.

Machine Learning Model to Predict Osteoporotic Spine with Hounsfield Units on Lumbar Computed Tomography.

ObjectiveBone mineral density (BMD) is an important consideration during fusion surgery. Although dual X-ray absorptiometry is considered as the gold standard for assessing BMD, quantitative computed tomography (QCT) provides more accurate data in spine osteoporosis. However, QCT has the disadvantage of additional radiation hazard and cost. The present study was to demonstrate the utility of artificial intelligence and machine learning algorithm for assessing osteoporosis using Hounsfield units (HU) of preoperative lumbar CT coupling with data of QCT. MethodsWe reviewed 70 patients undergoing both QCT and conventional lumbar CT for spine surgery. The T-scores of 198 lumbar vertebra was assessed in QCT and the HU of vertebral body at the same level were measured in conventional CT by the picture archiving and communication system (PACS) system. A multiple regression algorithm was applied to predict the T-score using three independent variables (age, sex, and HU of vertebral body on conventional CT) coupling with T-score of QCT. Next, a logistic regression algorithm was applied to predict osteoporotic or non-osteoporotic vertebra. The Tensor flow and Python were used as the machine learning tools. The Tensor flow user interface developed in our institute was used for easy code generation. ResultsThe predictive model with multiple regression algorithm estimated similar T-scores with data of QCT. HU demonstrates the similar results as QCT without the discordance in only one non-osteoporotic vertebra that indicated osteoporosis. From the training set, the predictive model classified the lumbar vertebra into two groups (osteoporotic vs. non-osteoporotic spine) with 88.0% accuracy. In a test set of 40 vertebrae, classification accuracy was 92.5% when the learning rate was 0.0001 (precision, 0.939; recall, 0.969; F1 score, 0.954; area under the curve, 0.900). ConclusionThis study is a simple machine learning model applicable in the spine research field. The machine learning model can predict the T-score and osteoporotic vertebrae solely by measuring the HU of conventional CT, and this would help spine surgeons not to under-estimate the osteoporotic spine preoperatively. If applied to a bigger data set, we believe the predictive accuracy of our model will further increase. We propose that machine learning is an important modality of the medical research field.

Multi-Dimensional Regression Host Utilization algorithm (MDRHU) for Host Overload Detection in Cloud Computing

The use of cloud computing data centers is growing rapidly to meet the tremendous increase in demand for high-performance computing (HPC), storage and networking resources for business and scientific applications. Virtual machine (VM) consolidation involves the live migration of VMs to run on fewer physical servers, and thus allowing more servers to be switched off or run on low-power mode, as to improve the energy consumption efficiency, operating cost and CO2 emission. A crucial step in VM consolidation is host overload detection, which attempts to predict whether or not a physical server will be oversubscribed with VMs. In contrast to the majority of previous work which use CPU utilization as the sole indicator for host overload, a recent study has proposed a multiple regression host overload detection algorithm, which takes multiple factors into consideration: CPU, memory and network BW utilization. This paper provides further improvement along two directions. First, we provide Multi-Dimensional Regression Host Utilization (MDRHU) algorithms that combine CPU, memory and network BW utilization via Euclidean Distance (MDRHU-ED) and absolute summation (MDRHU-AS), respectively. This leads to improved results in terms of energy consumption and service level agreement violation. Second, the study explicitly takes real-world HPC workloads into consideration. Our extensive simulation study further illustrates the superiority of our proposed algorithms over existing methods. In particular, as compared to the most recently proposed multiple regression algorithm that is based on Geometric Relation (GR), our proposed algorithms provide an improvement of at least 12% in energy consumption, and an improvement of at least 80% in a metric that combines energy consumption, service-level-violation, and number of VM migrations.

Regression Based ICT Model for Crop Yield Estimation

Farming and agriculture is the primary occupation in a nation like India. It has always proved to be important that we work towards constructing projects and systems that subsequently help in making a social reform where it is needed the most. Hence, it comes as no surprise that not only the government but also prominent science project centres have always paid a lot of attention and never hesitated from taking initiatives and efforts in assisting build different projects that help in development and progress in the same field. We also aim in building a project that draws an estimation about which crop is yielded at which rate in which district of Maharashtra. There are a number of algorithms that have proven to be helpful in finding out the estimation. We have taken use of Multiple Regression algorithm, which helped us to compute the yield estimation. The foremost goal of our project is to facilitate farmers and cultivators with an estimation system that helps them approximate the yield of crops and ultimately leads to a better and smarter farming structure for farm level.

Assessment of a Non-Optical Water Quality Property Using Space-Based Imagery in Egyptian Coastal Lake

Progressively anthropogenic intrusion and increasing water demand necessitate frequent water quality monitoring for sustainability management. Unlike laborious, time consuming field-based measurements, remote sensing-based water quality retrieval proved promising to overcome difficulties with temporal and spatial coverage. However, remotely estimated water quality parameters are mostly related to visibility characteristic and optically active property of water. This study presents results of an investigated approach to derive oxygen-related water quality parameter, namely Dissolved Oxygen (DO), in a shallow inland water body from satellite imagery. The approach deduces DO levels based on interrelated optical properties that dictate oxygen consumption and release in waters. Comparative analysis of multiple regression algorithms was carried out, using various combinations of parameters; namely, Turbidity, Total Suspended Solids (TSS), Chlorophyll-a, and Temperature. To cover the wide range of conditions that is experienced by Edku coastal lake, ground truth measurements covering the four seasons were used with corresponding satellite imageries. While results show successful statistically significant correlation in certain combinations considered, yet optimal results were concluded with Turbidity and natural logarithm of temperature. The algorithm model was developed with summer and fall data (R2 0.79), then validated with winter and spring data (R2 0.67). Retrieved DO concentrations highlighted the variability in pollution degree and zonation nature within that coastal lake, as related to boundary interactions and irregularity in flow dynamics within. The approach presented in this study encourages expanded applications with space-based earth observation products for exploring non-detectable water quality parameters that are interlinked with optically active properties in water.

Assessment of a Non-Optical Water Quality Property Using Space-based Imagery in Egyptian Coastal Lake

Abstract Progressive anthropogenic intrusion and increasing water demand necessitate frequent water quality monitoring for sustainability management. Unlike laborious, time consuming field-based measurements, remote sensing-based water quality retrieval proved promising to overcome difficulties with temporal and spatial coverage. However, remotely estimated water quality parameters are mostly related to visibility characteristic and optically active property of water. This study presents results of an investigated approach to derive oxygen –related water quality parameter, namely Dissolved Oxygen (DO), in a shallow inland water body from satellite imagery. The approach deduces DO levels based on interrelated optical properties that dictate oxygen consumption and release in waters. Comparative analysis of multiple regression algorithms was carried out, using various combinations of parameters; namely, Turbidity, Total Suspended Solids (TSS), Chlorophyll-a, and Temperature. To cover the wide range of conditions that is experienced by Edku coastal lake, ground truth measurements covering the four seasons were used with corresponding satellite imageries. While results show successful statistically significant correlation in certain combinations considered, yet optimal results were concluded with Turbidity and natural logarithm of temperature. The algorithm model was developed with summer and fall data (R2 0.79), then validated with winter and spring data (R2 0.67). Retrieved DO concentrations highlighted the variability in pollution degree and zonation nature within that coastal lake, as related to boundary interactions and irregularity in flow dynamics within. The approach presented in this study encourages expanded applications with space-based earth observation products for exploring non-detectable water quality parameters that are interlinked with optically active properties in water.

Remote sensing of CDOM absorption slope (S275–295) from satellite observations on the West Florida Shelf

Dissolved organic carbon (DOC) plays a critical role in climate-related biogeochemical cycles. The absorption spectrum slope of chromophoric dissolved organic matter (CDOM) in the 275–295 nm spectral range (S275–295) have been previously used as indicators of the source, molecular weight, photobleaching, and fate of DOC in estuarine and coastal environments. Here, we developed a multiple regression algorithm using field measurements and near-concurrent satellite observations in estimating S275–295 from satellites in the West Florida Shelf (WFS). The algorithm had a reliable performance in estimating S275–295 in terms of mean relative error and root mean square error < 10%. The algorithm was applied to available daily satellite images to establish long-term time series of S275–295 dynamics and to investigate its driving factors. S275–295 exhibited substantial spatiotemporal variability on the WFS. Correlation analyses revealed that river discharge regulated the temporal and spatial variation of S275–295 on the inner shelf, however, only wind had significant influence on the spatial and temporal distribution of S275–295 on the outer shelf. Climate variations and weather conditions had significant influences on the temporal and spatial distribution of S275–295 on the shelf by influencing the river discharge and wind at short time-scale.

Statistical Modeling via Bootstrapping and Weighted Techniques Based on Variances

Multiple logistic regression is a methodology of handling dependent variables with a binary outcome. This method is becoming increasingly widespread as a statistical technique that represents a discrete probability model. Many studies have focused on the application but less on the methodology building. This study aims to provide an applied method for multiple logistic regression which is called modified Bayesian logistic regression modeling as an alternative technique for logistic regression analysis that focuses on a combination of the bootstrap method using SAS macro and weighted techniques based on variances using SAS algorithm. Data on oral cancer were applied to illustrate a real scenario of oral health data. This data will be applied to the multiple logistic regression algorithm and modified Bayesian logistic regression. Results from both cases are strongly supported by clinical studies. Through the proposed algorithm, the researcher will have an option whether to analyze the data with the usual or an alternative method. Final results indicate that the modified procedure can provide more efficient results especially for the case which involves statistical inferences.

Statistical Modeling via Bootstrapping and Weighted Techniques Based on Variances

Multiple logistic regression is a methodology of handling dependent variables with a binary outcome. This method is becoming increasingly widespread as a statistical technique that represents a discrete probability model. Many studies have focused on the application but less on the methodology building. This study aims to provide an applied method for multiple logistic regression which is called modified Bayesian logistic regression modeling as an alternative technique for logistic regression analysis that focuses on a combination of the bootstrap method using SAS macro and weighted techniques based on variances using SAS algorithm. Data on oral cancer were applied to illustrate a real scenario of oral health data. This data will be applied to the multiple logistic regression algorithm and modified Bayesian logistic regression. Results from both cases are strongly supported by clinical studies. Through the proposed algorithm, the researcher will have an option whether to analyze the data with the usual or an alternative method. Final results indicate that the modified procedure can provide more efficient results especially for the case which involves statistical inferences.

Global Attenuation Relationship for Estimating Peak Ground Acceleration

Abstract Peak Ground Acceleration (PGA) is a very important ground motion parameter which is used to define the degree of ground shaking during an earthquake. It is also very helpful for designing earthquake resistant structure. The PGA can be estimated by attenuation relationships using magnitude, distance, source type etc of a ground motion. In the past, several researchers have developed over 450 attenuation relationships for predicting PGA for a specific region. In the present study an attempt has been made to develop an attenuation relationship on the basis of these available previous relationships in rock site which will be applicable for any region of the world. In the present study, PGA has been expressed as a function of moment magnitude and hypo-central distance in rock site. Chi-square test have also been performed with available earthquake data in American and Indian region for verifying the accuracy of the generated attenuation relationship. Using multiple regression and Genetic Algorithm (GA) the attenuation relationship equations have also been generated. These equations will be very helpful for performing seismic hazard analysis and predicting earthquake force in any region of the world.

Impact of grain morphology and the genotype by environment interactions on test weight of spring and winter wheat (Triticum aestivum L.)

Wheat (Triticum aestivum L.) market grades and prices are determined in part by test weight (TW). Millers value high TW because it is typically associated with higher flour extraction rates and better end-use quality. Test weight is expected to be influenced by other directly quantifiable grain attributes such as grain length (GL), grain width (GW), shape, single-grain-density (SGD), thousand-grain-weight (TGW), and packing efficiency (PE). The objectives of this study were to: (1) determine the primary morphological grain attributes that comprise TW measurements for winter and spring wheat classes; and (2) determine TW stability and genotype and genotype × environment interactions (GEIs) of the attributes that comprise TW. A market class representative group of 32 hard spring and 24 hard winter wheat cultivars was grown at several locations in South Dakota in 2011 and 2012. A regularized multiple regression algorithm was used to develop a TW model and determine what grain attribute reliably predicts TW. A GGE biplot was used for stability and GEI analyses whereas a linear mixed model was used for variance analyses. Data were collected on eight grain traits: TW, SGD, TGW, protein concentration, GW, GL, shape, size, and PE. Observations showed that in both spring and winter wheat, SGD accounted for over 90% of the phenotypic variation of TW. Cultivars with stable and high TW were identified in both wheat classes. Apart from TW; significant (p < 0.0001) genotype, environment, and GEI variances were observed for GW and SGD, a more direct measure of which could help improve genetic gain for TW.

The Prediction Model of Cotton Yarn Intensity Based on the CNN-BP Neural Network

Yarn strength index is a heavy index of yarn quality, Yarn quality can be well controlled by predicting yarn strength index. Generally, multiple non regression algorithms, support vector machines (SVD) and BP neural network algorithms are generally used to predict yarn strength. This paper presents an algorithm to connect the convolution neural network (CNN) with the BP neural network, which is written as the CNN-BP algorithm. We use 20 sets of data to train CNN-BP algorithm, regression, V-SVD algorithm, and BP neural network. We tested CNN-BP algorithm, regression, V-SVD algorithm, and BP neural network with 5 sets of data. The CNN-BP neural network algorithm is the best in these four algorithms.

Can Empirical Algorithms Successfully Estimate Aragonite Saturation State in the Subpolar North Atlantic?

The aragonite saturation state (ΩAr) in the subpolar North Atlantic was derived using new regional empirical algorithms. These multiple regression algorithms were developed using the bin-averaged GLODAPv2 data of commonly observed oceanographic variables (temperature (T), salinity (S), pressure (P), oxygen (O2), nitrate (NO3-), phosphate (PO4-3), silicate (Si(OH)4), and pH). Five of these variables are also frequently observed using autonomous platforms, which means they are widely available. The algorithms were validated against independent shipboard data from the OVIDE2012 cruise. It was also applied to time series observations of T, S, P and O2 from the K1 mooring (56.5°N, 52.6°W) to reconstruct for the first time the seasonal variability of ΩAr. Our study suggests: (i) linear regression algorithms based on bin-averaged carbonate system data can successfully estimate ΩAr in our study domain over the 0-3500m depth range (R2=0.985, RMSE= 0.044); (ii) that ΩAr also can be adequately estimated from solely non-carbonate observations (R2=0.969, RMSE=0.063) and autonomous sensor variables (R2=0.978, RMSE=0.053). Validation with independent OVIDE2012 data further suggests that (iii) both algorithms, non-carbonate (MEF=0.929) and autonomous sensors (MEF=0.995) have excellent predictive skill over the 0-3500 depth range; (iv) that in deep waters (>500m) observations of T, S and O2 may be sufficient predictors of ΩAr (MEF=0.913); (iv) the importance of adding pH sensors on autonomous platforms in the euphotic and remineralization zone (<500m). Reconstructed ΩAr at Irminger Sea site, and the K1 mooring in Labrador Sea show high seasonal variability at the surface due to biological drawdown of inorganic carbon during the summer, and fairly uniform ΩAr values in the water column during winter convection. Application to time series sites shows the potential for regionally tuned algorithms, but they need to be further compared against ΩAr calculated by conventional means to fully assess their validity and performance.