Combination Of Explanatory Variables Research Articles

Reconceptualizing threshold‐mediated runoff responses: A case study from the Humber River watershed, Ontario, Canada

AbstractWatershed‐scale runoff responses are driven by various factors including climate, geology, soils, topography and landcover. They are often threshold‐mediated, expressing significant changes in hydrologic behaviour at critical moments in time or points in space. The influence of multiple explanatory variables on rainfall‐runoff relationships is not adequately captured by commonly applied approaches portraying runoff responses as a function of one variable related to watershed storage. In this case study, a novel approach was borrowed from ecological research to quantify and better understand threshold‐mediated runoff responses. Modelled three‐dimensional surfaces depicting metrics of event runoff responses as a function of rainfall amount and intensity were analysed to quantify both the abruptness of potential thresholds (i.e., threshold strength) and the simultaneous influence of different rainfall characteristics on the response (i.e., diagonality). The approach was applied to sub‐watersheds of the Humber River (Ontario, Canada), which have a nested configuration and a strong land use gradient, providing an opportunity to explore how the interplay between rainfall amount and intensity in determining runoff response is affected by sub‐watershed physical features. The study revealed that threshold strengths and the simultaneous influence of rainfall amount and intensity varied, depending on the sub‐watershed and event‐specific conditions. There was evidence that sub‐watershed slope and imperviousness along with the watershed position relative to prevailing weather patterns influences threshold strength and diagonality. This research extends threshold analyses in hydrology to encompass multiple explanatory variables: it aligns more closely with perceptual models of runoff generation and encourages a reimagining of thresholds as discontinuities in response across various combinations of explanatory variables. The threshold strength and diagonality parameters facilitate objective comparisons of thresholds across space and time and may be valuable tools for watershed classification and inter‐comparison, and for evaluating and/or calibrating rainfall‐runoff models. These promising lines of inquiry would be best served by applying this methodology across a broader range of spatial scales and hydroclimatic conditions.

Searching for the Inflection Point of Travel Well-Being from the Views of Travel Characteristics Based on the Ordered Logistic Regression Model

Travel well-being is the subjective feeling of satisfaction that people have while traveling. Previous research focused on its determinants and relationships with subjective well-being ignored. But no quantitative study discusses the effect of characteristics like weekly income and travel time on travel well-being. To demonstrate the quantitative inflection of travel well-being from characteristics, the relevant factors influencing travel well-being as the dependent variable are identified using Pearson correlation analysis and linear regression in this paper. To overcome the limitations of linear regression techniques, ordered logistic regression is applied to establish an analytical model of travel well-being for predicting the response probabilities for different degrees based on combinations of explanatory variables. Both the linear regression and ordered logistic regression models are calibrated by American residents’ travel datasets. The results illustrate that the ordered logistic model fits sample data better than linear regression models. Age, travel time, health status, and resting degree are significantly related to travel well-being. Older people and those who are healthier and better rested are more likely to experience higher levels of travel well-being. Additionally, increased travel time is associated with a significant decrease in travel well-being. Therefore, to enhance people’s travel feelings, policymakers and urban planners can enhance the quality of public transportation services and provide diverse transportation options, while also logically constructing transportation hubs to provide more convenient travel plans.

Spatial estimation and yearly trends in abundance-index of Japanese jack mackerel (Trachurus japonicus) in the East China Sea and Sea of Japan

Catch per unit effort (CPUE) of commercial fisheries is recommended to be standardized because various factors can affect catch rates. Recently, the spatio-temporal analysis has been required to standardize fishery data, because catch is spatially distributed in proportion to fishery distribution and not in proportion to fish abundance. We developed the spatio-temporal models for the CPUE of large- and medium-sized purse seine fisheries for the different age groups of Japanese jack mackerel (Trachurus japonicus) to evaluate the potential spatial distribution and estimate the annual trends in abundance index. Japanese jack mackerel is widely distributed in the East Asian seas, including the East China Sea and Sea of Japan, and is an important fishery resource in Japan. Our models were delta-lognormal generalized additive mixed models that incorporated the temporal components of year and month, and size of fishing vessels as categorical variables, and the identity of the fishing vessels as a random variable. The spatial component represented by the latitude and longitude points of the fishing location was smoothed using an isotropic bivariate function. The water temperature at a depth of 50 m was incorporated to detect the effects of environmental factors. After all combinations of explanatory variables were evaluated, the model with the lowest Bayesian information criterion score was selected as the best model. All the best models among both binomial models and lognormal models contained all explanatory variables, although the best binomial model for age 2 exclude the vessel size and the best lognormal model for age 0 exclude the water temperature at a depth of 50 m. Our results indicate that the potential distribution areas for all ages of Japanese jack mackerel exist on the continental shelf in the southern part of the East China Sea, the western side of Kyushu Island, and along Honshu Island in the western part of the Sea of Japan. In contrast, the main distribution areas with higher density for age 0 were segregated into two: the southwestern East China Sea and the coastal area off Kyushu Island, although fish over age 1 were widely distributed. The standardized CPUE was higher than the nominal CPUE for age 0 from 2015 to 2020, and the standardized CPUE was also higher for age 1 in 2020. These results were derived from spatial models that compensated for the lack of spatial data due to the recent decrease in fishing operations in the Southeast China Sea. The yearly trends of standardized CPUE for each age could be used for the age-structured stock assessment model such as a tuned virtual population analysis. Improving the accuracy of the abundance index helps to further improve the reliability of stock assessments.

An empirical assessment of the tripartite nexus between environmental pollution, economic growth, and agricultural production in Sub-Saharan African countries.

A lot of attention has been paid to environmental pollution worldwide, due to the increase in anthropogenic activities. Massive investment in non-renewable energy options raises questions regarding environmental sustainability and how to maximize food and non-food output while still preserving a healthy ecosystem. To this end, the present study explores the three-way nexus between economic growth, CO2 emission, and agriculture-value added will accounting for other control variables across a balanced panel of selected African economies from 1997 to 2020. Panel econometrics method of the generalized method of moments (two-step difference GMM) is used to obtain a robust result. From the present study, the environmental pollution model shows that economic growth significantly contributes to environmental pollution in Africa. Additionally, the food price index, capital, and FDI promote pollution, while agricultural production and labor decrease pollution. In the case of the economic growth model, the findings reveal that environmental pollution supports the growth-led pollution hypothesis. Also, the food price index and capital ameliorate economic growth, while foreign direct investments decrease economic growth. Finally, the agricultural production model indicates that economic growth increases agricultural production when the interaction term between GDPC and FDI is included in the model. In summary, the combination of explanatory variables, environmental pollution, capital, and foreign direct investment decreases agricultural production. On the contrary, the food price index and labor promote agricultural production in Africa. Furthermore, the study provides a lot of policies for authorities and stakeholders in Sub-Saharan African countries and other developing economies.

Constructing the composition ratio prediction model using machine learning for BaSi2 thin films deposited by thermal evaporation

A composition ratio prediction model for BaSi2 thin films deposited by thermal evaporation was constructed using machine learning. BaSi2 was prepared by thermal evaporation in a vacuum chamber, and the composition ratio was measured by energy-dispersive X-ray spectroscopy. The results show that the composition ratio is affected by various experimental parameters. To consider these parameters, kernel ridge regression was performed with Si/Ba ratio as the objective variable, and with experimental parameters as explanatory variables. A good fitting result was obtained by kernel ridge regression. The next step was to select a kernel function. We evaluated four types of kernel functions, and confirmed that two of them, the polynomial kernel and the sigmoid kernel, have relatively high prediction accuracy. Then we investigated different combinations of explanatory variables and found the best combination with the highest generalization performance. From the above, a composition ratio prediction model with a mean absolute error of less than 0.2 was obtained.

Leveraging Important Covariate Groups for Corn Yield Prediction

Accurate yield information empowers farmers to adapt, their governments to adopt timely agricultural and food policy interventions, and the markets they supply to prepare for production shifts. Unfortunately, the most representative yield data in the US, provided by the US Department of Agriculture, National Agricultural Statistics Service (USDA-NASS) Surveys, are spatiotemporally patchy and inconsistent. This paper builds a more complete data product by examining the spatiotemporal efficacy of random forests (RF) in predicting county-level yields of corn—the most widely cultivated crop in the US. To meet our objective, we compare RF cross-validated prediction accuracy using several combinations of explanatory variables. We also utilize variable importance measures and partial dependence plots to compare and contextualize how key variables interact with corn yield. Results suggest that RF predicts US corn yields well using a relatively small subset of climate variables along with year and geographical location (RMSE = 17.1 bushels/acre (1.2 tons/hectare)). Of note is the insensitivity of RF prediction accuracy when removing variables traditionally thought to be predictive of yield or variables flagged as important by RF variable importance measures. Understanding what variables are needed to accurately predict corn yields provides a template for applying machine learning approaches to estimate county-level yields for other US crops.

Examining the role of financial innovation on economic growth: Fresh empirical evidence from developing and developed countries

From an economic prospective, the relationship between financial innovation and economic growth has always been a critical aspect of the financial system. However, this relationship has not recieved extensive attention in academia. Therefore, this study examines the effect of financial innovation on economic growth in the context of developing and developed countries using 92 cross-sectional data over the period of 2002 to 2020, the fixed effect method of panel regression has been applied to examine the relationship between financial innovation and economic growth. This study has compared the results of developing and developed countries using different combinations of explanatory variables. The results of this study depict that various combinations of variables have varied impacts on dependent variable. This study has revealed the negative relationship between financial innovation and economic growth. However, domestic credit to private sector banks has the best proxy for financial innovation. This study suggests that regulators should focus on encouraging and discouraging financial innovation based on its impact on the economy.

A probabilistic framework to model bond strength between concrete and GFRP rebar using Bayesian linear regression

This study was aimed at developing a probabilistic model using Bayesian linear regression approach to examine bond strength behavior between concrete and GFRP rebars. This probabilistic model presents a quantitative description of the epistemic uncertainties of model parameters and model error, which decrease as new information is received. By using the problem mechanics, variables were selected and explanatory functions were developed. Then, based on an observational database including 405 beam test specimens, 47 candidate models were selected based on different combinations of explanatory variables, and the fittest model was selected based on prediction quality, error non-normality, scattering non-homogeneity, residual correlation, and nonlinearity. Afterward, Bayesian regression was used to obtain the probability distribution of the model parameters, and model error was obtained using Bayesian regression. The paper continues by an in-depth explanation of the observation selection process, model development, model diagnostic, step-by-step model reduction process with the removal of ineffective parameters. The analysis results specified the bar position and transverse reinforcement ratio parameters as ineffective parameters. In addition, a comparison between deterministic and probabilistic models showed that contrary to the deterministic model, the probabilistic one could capture uncertainty in the model.

Susceptibility Assessment of Groundwater Nitrate Contamination Using an Ensemble Machine Learning Approach.

Groundwater pollution susceptibility mapping using parsimonious approaches with limited data is of utmost importance for water resource and health planning, especially in data-scarce regions. Current research assesses groundwater nitrate susceptibility by considering the various combination of explanatory variables. In this study, the novel machine learning models of weighted subspace random forest (WSRF) and generalized additive model using LOESS (GAMLOESS) are applied, and the results are compared with well-known machine learning models of K-nearest neighbors (KKNN) and random forest (RF). The optimum combination of inputs for groundwater nitrate susceptibility mapping is identified using the k-fold cross-validation methodology. Results indicated that the combination of variables of precipitation, groundwater level, and lithology had the best performance among the 16 combinations. Modeling performance using the optimum combination demonstrated that the new ensemble approach, the WSRF model, had superior performance according to the evaluation metrics of accuracy (0.87), kappa (0.73), precision (0.92), false alarm ratio (0.08), and critical success index (0.75). The susceptibility assessment results of this paper can be a useful tool in developing strategies for the prevention and protection of groundwater pollution.

Online job vacancy attractiveness: Increasing views, reactions and conversions

E-commerce development has not bypassed the labor market, and employers seeking to hire are posting vacancies on specialized online platforms. Operators of such online platforms and employers are assessing the attractiveness of job vacancies, as they are interested in making online job vacancies (OJVs) more attractive to job seekers, i.e. to increase the number of views of the posted job ad, the number of times job seekers submit an application form and conversions (the ratio of the two). However, given that job types and job seekers are extremely heterogeneous, it is difficult for employers and specialized online platform operators to design attractive job ads. In collaboration with a leading OJV platform in Slovakia, we study whether machine-learning methods can improve the predictions of OJV attractiveness on a sample of 32482 OJVs that offer as many as 883 job features. Our study shows that, as opposed to various linear models, considerable prediction improvements can be achieved using random forest, which exploits possible nonlinear relationships between combinations of explanatory variables and measures of attractiveness. Based on this insight, we perform a statistical evaluation of key variables of importance. We find that the job classification, job benefits, and variables related to a simple morphological description of the job and job title are relevant. The results of this study can help the operators of specialized job vacancy platforms and employers improve their job ads to attract more job seekers.

An investigation of penalization and data augmentation to improve convergence of generalized estimating equations for clustered binary outcomes

BackgroundIn binary logistic regression data are ‘separable’ if there exists a linear combination of explanatory variables which perfectly predicts the observed outcome, leading to non-existence of some of the maximum likelihood coefficient estimates. A popular solution to obtain finite estimates even with separable data is Firth’s logistic regression (FL), which was originally proposed to reduce the bias in coefficient estimates. The question of convergence becomes more involved when analyzing clustered data as frequently encountered in clinical research, e.g. data collected in several study centers or when individuals contribute multiple observations, using marginal logistic regression models fitted by generalized estimating equations (GEE). From our experience we suspect that separable data are a sufficient, but not a necessary condition for non-convergence of GEE. Thus, we expect that generalizations of approaches that can handle separable uncorrelated data may reduce but not fully remove the non-convergence issues of GEE.MethodsWe investigate one recently proposed and two new extensions of FL to GEE. With ‘penalized GEE’ the GEE are treated as score equations, i.e. as derivatives of a log-likelihood set to zero, which are then modified as in FL. We introduce two approaches motivated by the equivalence of FL and maximum likelihood estimation with iteratively augmented data. Specifically, we consider fully iterated and single-step versions of this ‘augmented GEE’ approach. We compare the three approaches with respect to convergence behavior, practical applicability and performance using simulated data and a real data example.ResultsOur simulations indicate that all three extensions of FL to GEE substantially improve convergence compared to ordinary GEE, while showing a similar or even better performance in terms of accuracy of coefficient estimates and predictions. Penalized GEE often slightly outperforms the augmented GEE approaches, but this comes at the cost of a higher burden of implementation.ConclusionsWhen fitting marginal logistic regression models using GEE on sparse data we recommend to apply penalized GEE if one has access to a suitable software implementation and single-step augmented GEE otherwise.

National Intelligence and Economic Growth: A Bayesian Update

Since Lynn and Vanhanen’s book IQ and the Wealth of Nations (2002), many publications have evidenced a relationship between national IQ and national prosperity. The strongest statistical case for this lies in Jones and Schneider’s (2006) use of Bayesian model averaging to run thousands of regressions on GDP growth (1960-1996), using different combinations of explanatory variables. This generated a weighted average over many regressions to create estimates robust to the problem of model uncertainty. We replicate and extend Jones and Schneider’s work with many new robustness tests, including new variables, different time periods, different priors and different estimates of average national intelligence. We find national IQ to be the “best predictor” of economic growth, with a higher average coefficient and average posterior inclusion probability than all other tested variables (over 67) in every test run. Our best estimates find a one point increase in IQ is associated with a 7.8% increase in GDP per capita, above Jones and Schneider’s estimate of 6.1%. We tested the causality of national IQs using three different instrumental variables: cranial capacity, ancestry-adjusted UV radiation, and 19th-century numeracy scores. We found little evidence for reverse causation, with only ancestry-adjusted UV radiation passing the Wu-Hausman test (p < .05) when the logarithm of GDP per capita in 1960 was used as the only control variable. Key Words: Human capital, National IQ, Economic growth, Bayesian model averaging, Intelligence, Smart fraction

Mobile monitoring and spatial predictionof black carbonin Cairo, Egypt.

This study harnessesthe power of mobile data in developing a spatialmodel for predicting black carbon (BC) concentrations within one of the most heavily populated regions in the Middle East and North Africa MENA region, Greater Cairo Region (GCR) in Egypt. A mobile data collection campaign was conducted in GCR to collect BC measurements along specific travel routes. In total, 3,300km were travelled across a widespread 525km of routes. Reported average BC values were around 20µg/m3, announcing an alarming order of magnitude value when compared to the maximum reported values in similar studies. A bi-directional stepwise land use regression (LUR) model was developed to select the best combination of explanatory variables and generate an exposure surface for BC, in addition to a number of machine learning models (random forest gradient boost, light gradient boost model (LightGBM), Keras neural network (NN)). Data from 7 air quality (AQ) stations were compared-in terms of mean square error (MSE) and mean absolute error (MAE)-with predictions from the LUR and the NN model. The NN model estimated higher BC concentrations in the downtown areas, while lower concentrations are estimated for the peripheral area at the east side of the city. Such results shed light on the credibility of the LUR models in generating a general spatial trend of BC concentrations while the superiority of NN in BC accuracy estimation (0.023 vs 0.241 in terms of MSE and 0.12 vs 0.389 in terms of MAE; of NN vs LUR respectively).

Mapping structural attributes of tropical dry forests by combining Synthetic Aperture Radar and high‐resolution satellite imagery data

AbstractAimOptical satellite imagery has been used for mapping the spatial distribution of vegetation structure attributes; however, obtaining accurate estimates with optical imagery can be difficult in tropical forests due to their dense canopy and multi‐layered vegetation. Synthetic aperture radar imagery can be more suitable in this case, as the radar signal can penetrate the forest canopy and interact with stems, providing a better estimation of the vegetation structure. This study compared the accuracy of forest species richness, tree diameter, height, and basal area estimates obtained using Sentinel‐2 and Advanced Land Observing Satellite ‐1 (ALOS) Phased Array type L‐band Synthetic Aperture Radar (PALSAR) data, either combined or separately.LocationThe Yucatan Peninsula, Mexico.MethodsField data were collected in three 3600‐km2‐window areas with three different types of tropical dry forest. Three random forest regression models were fitted: one using explanatory variables derived from Sentinel‐2 data, a second using predictor variables derived from ALOS PALSAR, and the third using a combination of explanatory variables from both sensors. A variance partitioning analysis was carried out to examine the percent variability of each vegetation attribute that was explained by the models combining the explanatory variables of the two sensors (ALOS PALSAR and Sentinel‐2).ResultsVegetation attribute estimation errors ranged from 13% to 38.5% when using ALOS PALSAR variables and from 11% to 33% when using Sentinel‐2 variables. Combining variables from both sensors provided more accurate estimates of vegetation attributes. A 5% reduction of the estimated error, and an increase from 0.50 to 0.63 of the percentage of variation explained by the models (R2) were achieved.ConclusionsOur results suggest that both ALOS PALSAR and Sentinel‐2 data provide accurate estimates of vegetation structure and species richness in tropical dry forests. However, combining explanatory variables from the two sensors improved the estimation accuracy of vegetation attributes.

Performances of some high dimensional regression methods: sparse principal component regression

Principal component analysis (PCA) is widely used technique in data processing and dimensionality reduction, but it has some drawbacks since each principal component is a linear combination of explanatory variables. As an alternative, the sparse PCA (SPCA) is a very appealing method which produces principal components with sparse loadings. On the other hand, combining PCA on explanatory variables with least squares regression yields to principal component regression (PCR). In PCR, the components are obtained using only explanatory variables, not considered the effect of the dependent variable. Considering the dependent variable, the sparse PCR (SPCR) enables to obtain sparse principal component loadings. But the main drawback of it is the computational cost. Taking into consideration the general structure of PCR, we combine (S)PCA with some sparse regression methods and compared with the classical PCR and last introduced method, SPCR. Extensive simulation studies and real data examples are implemented to show their performances. The results are supported by a reasonable computation time study.

Real Time Safety Model for Pedestrian Red-Light Running at Signalized Intersections in China

The traditional way to evaluate pedestrian safety is a reactive approach using the data at an aggregate level. The objective of this study is to develop real-time safety models for pedestrian red-light running using the signal cycle level traffic data. Traffic data for 464 signal cycles during 16 h were collected at eight crosswalks on two intersections in the city of Nanjing, China. Various real-time safety models of pedestrian red-light running were developed based on the different combination of explanatory variables using the Bayesian Poisson-lognormal (PLN) model. The Bayesian estimation approach based on Markov chain Monte Carlo simulation is utilized for the real-time safety models estimates. The models’ comparison results show that the model incorporated exposure, pedestrians’ characteristics and crossing maneuver, and traffic control and crosswalk design outperforms the model incorporated exposure and the model incorporated exposure, pedestrians’ characteristics, and crossing maneuver. The result indicates that including more variables in the real-time safety model could improve the model fit. The model estimation results show that pedestrian volume, ratio of males, ratio of pedestrians on phone talking, pedestrian waiting time, green ratio, signal type, and length of crosswalk are statistically significantly associated with the pedestrians’ red-light running. The findings from this study could be useful in real-time pedestrian safety evaluation as well as in crosswalk design and pedestrian signal optimization.

Afforestation driving long-term surface water browning.

Increase in surface water color (browning), caused by rising dissolved organic carbon (DOC) and iron concentrations, has been widely reported and studied in the last couple of decades. This phenomenon has implications to aquatic ecosystem function and biogeochemical carbon cycling. While recovery from acidification and changes in climate‐related variables, such as precipitation and length of growing season, are recognized as drivers behind browning, land‐use change has received less attention. In this study, we include all of the above factors and aim to discern their individual and combined contribution to water color variation in an unprecedentedly long (1940–2016) and highly resolved dataset (~20 times per month), from a river in southern Sweden. Water color showed high seasonal variability and a marked long‐term increase, particularly in the latter half of the dataset (~1980). Short‐term and seasonal variations were best explained by precipitation, with temperature playing a secondary role. All explanatory variables (precipitation, temperature, S deposition, and land‐use change) contributed significantly and together predicted 75% of the long‐term variation in water color. Long‐term change was best explained by a pronounced increase in Norway spruce (Picea abies Karst) volume—a measure of land‐use change and a proxy for buildup of organic soil layers—and by change in atmospheric S deposition. When modeling water color with a combination of explanatory variables, Norway spruce showed the highest contribution to explaining long‐term variability. This study highlights the importance of considering land‐use change as a factor behind browning and combining multiple factors when making predictions in water color and DOC.

A Multivariate Negative-Binomial Model with Random Effects for Differential Gene-Expression Analysis of Correlated mRNA Sequencing Data.

Experimental designs such as matched-pair or longitudinal studies yield mRNA sequencing (mRNA-Seq) counts that are correlated across samples. Most of the approaches for the analysis of correlated mRNA-Seq data are restricted to a specific design and/or balanced data only (with the same number of samples in each group). We propose a model that is applicable to the analysis of correlated mRNA-Seq data of different types: paired, clustered, longitudinal, or others. Any combination of explanatory variables, as well as unbalanced data, can be processed within the proposed modeling framework. The model assumes that exon counts of a particular gene of an individual sample jointly follow a multivariate negative-binomial distribution. Additional correlation between exon counts obtained for, for example, individual samples within the same pair or cluster, is taken into account by including into the model a cluster-level normally distributed random effect. An interesting feature of the model is that it provides explicit expression for marginal correlation between exon counts at different levels. The performance of the model is evaluated by using a simulation study and an analysis of two real-life data sets: a paired mRNA-Seq experiment for 24 patients with clear-cell renal-cell carcinoma and a longitudinal mRNA-Seq experiment for 29 patients with Lyme disease.

Application of continuous turbidity sensors to supplement estimates of total phosphorus concentrations in the Grand River, Ontario, Canada

Accurate estimates of total phosphorus (TP) loadings to eastern Lake Erie are critical for developing load reduction targets and for determining if commitments are being met under the Great Lakes Water Quality Agreement, 2012 (GLWQA). Currently, loading calculations from Canadian priority tributaries are supported by year-round event-focused water quality sampling using automated samplers and laboratory water quality measurements. Here we evaluate the suitability of continuously-measured parameters, namely turbidity and flow, to supplement or enhance knowledge about TP concentrations in the Grand River, ON, by providing continuous data alongside event-focused sample measurements. A series of simple and multiple linear regression models were evaluated and compared with respect to their ability to predict TP water concentrations as a function of different combinations of explanatory variables. Explanatory variables included turbidity, flow, season and flow condition (i.e. hysteresis). The models that performed best explained 63–65% of the variation of TP which is comparable to surrogate model applications in the U. S and elsewhere. Additional model calibration work is needed due to gaps in turbidity data particularly during high flow events. We emphasize the need for continued automated, event-focused water quality sampling. However, provided that operational challenges are overcome, our results indicate that sensor-derived water quality parameters to predict TP concentrations is a promising technique that may supplement and improve nutrient loading estimates in the Grand River into the future and provides guidance for the utilization of this method in other tributaries.

Drivers of meat consumption

Increasing global levels of meat consumption are a threat to the environment and to human health. To identify measures that may change consumption patterns towards more plant-based foods, it is necessary to improve our understanding of the causes behind the demand for meat. In this paper we use data from 137 different countries to identify and assess factors that influence meat consumption at the national level using a cross-country multivariate regression analysis. We specify either total meat or ruminant meat as the dependent variable and we consider a broad range of potential drivers of meat consumption. The combination of explanatory variables we use is new for this type of analysis. In addition, we estimate the relative importance of the different drivers. We find that income per capita followed by rate of urbanisation are the two most important drivers of total meat consumption per capita. Income per capita and natural endowment factors are major drivers of ruminant meat consumption per capita. Other drivers are Western culture, Muslim religion, female labour participation, economic and social globalisation and meat prices. The main identified drivers of meat demand are difficult to influence through direct policy intervention. Thus, acting indirectly on consumers’ preferences and consumption habits (for instance through information, education policy and increased availability of ready-made plant based products) could be of key importance for mitigating the rise of meat consumption per capita all over the world.