How to find “empty” buffer?

So I have lines(road network) and buffers around them. I need to find "empty bufers" whose lines is missing (pic below). Lines have attributes of buffer width, but it is incorrect in some cases (does not match with actual buffer with). So creating/erasing with new polygons is not an option, since it will leave alot of trash polygons. Buffers is also dissolved in most cases.

This is just an idea it would need further experimenting as it may be flawed?

How about looking at some sort of ratio value of length/perimeter?

Imagine you had a line that is 10m long and has its full buffer which you had buffered out by 2m then the perimeter of the buffer would be approximately 20m. so 10/20 = 0.5.

Now imagine you had a line that was 25m long in an L shaped buffer that had been created by a 50m L shaped polyline for which you had only one side of it so giving your scenario of a half empty buffer. You would have a 25m / ~100m perimeter = 0.25.

So values close to 0.5 are polygons with their central lines, would need to test what a good cut off value would be.

Just an idea…

This is just another idea, by looking at the opposite angle to your question, to find centrelines first as explained one of the methods in here and then search for matching line features. Of course this approach will give robust results if the road lines are at the 'centre' of the polygons. Maybe adding thin ('how thin' is another issue of course) buffers to your existing road features, which will introduce a pseudo-tolerance, help you to find similar features by looking at how much of the centreline is 'contained' by the buffer.

Background: Studies estimating health effects of long-term air pollution exposure often use a two-stage approach: building exposure models to assign individual-level exposures, which are then used in regression analyses. This requires accurate exposure modeling and careful treatment of exposure measurement error.

Objective: To illustrate the importance of accounting for exposure model characteristics in two-stage air pollution studies, we considered a case study based on data from the Multi-Ethnic Study of Atherosclerosis (MESA).

Methods: We built national spatial exposure models that used partial least squares and universal kriging to estimate annual average concentrations of four PM2.5 components: elemental carbon (EC), organic carbon (OC), silicon (Si), and sulfur (S). We predicted PM2.5 component exposures for the MESA cohort and estimated cross-sectional associations with carotid intima-media thickness (CIMT), adjusting for subject-specific covariates. We corrected for measurement error using recently developed methods that account for the spatial structure of predicted exposures.

Results: Our models performed well, with cross-validated R 2 values ranging from 0.62 to 0.95. Naïve analyses that did not account for measurement error indicated statistically significant associations between CIMT and exposure to OC, Si, and S. EC and OC exhibited little spatial correlation, and the corrected inference was unchanged from the naïve analysis. The Si and S exposure surfaces displayed notable spatial correlation, resulting in corrected confidence intervals (CIs) that were 50% wider than the naïve CIs, but that were still statistically significant.

Conclusion: The impact of correcting for measurement error on health effect inference is concordant with the degree of spatial correlation in the exposure surfaces. Exposure model characteristics must be considered when performing two-stage air pollution epidemiologic analyses because naïve health effect inference may be inappropriate.

Citation: Bergen S, Sheppard L, Sampson PD, Kim SY, Richards M, Vedal S, Kaufman JD, Szpiro AA. 2013. A national prediction model for PM2.5 component exposures and measurement error–corrected health effect inference. Environ Health Perspect 121:1017–1025 http://dx.doi.org/10.1289/ehp.1206010

Introduction

The relationship between air pollution and adverse health outcomes has been well documented (Pope et al. 2002 Samet et al. 2000). Many studies focus on particulate matter, specifically particulate matter ≤ 2.5 μm in aerodynamic diameter (PM2.5) (Kim et al. 2009 Miller et al. 2007). Health effects of PM2.5 may depend on characteristics of the particles, including shape, solubility, pH, or chemical composition (Vedal et al., in press), and a deeper understanding of these differential effects could help inform policy. One of the challenges in assessing the impact of different chemical components of PM2.5 in an epidemiologic study is the need to assign exposures to study participants based on monitoring data from different locations (i.e., spatially misaligned data). When doing this for many components, the prediction procedure needs to be streamlined in order to be practical. Whatever the prediction algorithm, using the estimated rather than true exposures induces measurement error in the subsequent epidemiologic analysis. Here we describe a flexible and efficient prediction model that can be applied on a national scale to estimate long-term exposure levels for multiple pollutants and that implements existing methods of correcting for measurement error in the health model.

Current methods for assigning exposures include land-use regression (LUR) with geographic information system (GIS) covariates (Hoek et al. 2008) and universal kriging, which also exploits residual spatial structure (Kim et al. 2009 Mercer et al. 2011). Often hundreds of candidate correlated GIS covariates are available, necessitating a dimension reduction procedure. Variable selection methods that have been considered in the literature include exhaustive search, stepwise selection, and shrinkage by the “lasso” (Mercer et al. 2011 Tibshirani 1996). However, variable selection methods tend to be computationally intensive, feasible perhaps when considering a single pollutant but quickly becoming impractical when developing predictions for multiple pollutants. A more streamlined alternative is partial least squares (PLS) regression (Sampson et al. 2009), which finds a small number of linear combinations of the GIS covariates that most efficiently account for variability in the measured concentrations. These linear combinations reduce the covariate space to a much smaller dimension and can then be used as the mean structure in a LUR or universal kriging model in place of individual GIS covariates. This provides the advantages of using all available GIS covariates and eliminating potentially time-consuming variable selection processes.

Using exposures predicted from spatially misaligned data rather than true exposures in health models introduces measurement error that may have implications for ^ βx, the estimated health model coefficient of interest (Szpiro et al. 2011b). Berkson-like error that arises from smoothing the true exposure surface may inflate the SE of ^ βx. Classical-like error results from estimating the prediction model parameters and may bias ^ βx in addition to inflating its SE. Bootstrap methods to adjust for the effects of measurement error have been discussed by Szpiro et al. (2011b).

Here we present a case study to illustrate a holistic approach to two-stage air pollution epidemiologic modeling, which includes exposure modeling in the first stage and health modeling that incorporates measurement error correction in the second stage. We build national exposure models using PLS and universal kriging, and employ them to estimate long-term average concentrations of four chemical species of PM2.5—elemental carbon (EC), organic carbon (OC), silicon (Si), and sulfur (S)—selected to reflect a variety of different PM2.5 sources and formation processes (Vedal et al., in press). After developing the exposure models, we derive predictions for the Multi-Ethnic Study of Atherosclerosis (MESA) cohort. These predictions are used as the covariates of interest in health analyses assessing associations between carotid intima-media thickness (CIMT), a subclinical measure of atherosclerosis, and exposure to PM2.5 components. We apply measurement error correction methods to account for the fact that predicted rather than true exposures are being used in these health models. We discuss our results and their implications with regard to the effect of spatial correlation in exposure surfaces on estimated associations between exposures and health outcomes.

Monitoring data. Data on EC, OC, Si, and S were collected to build the national models. These data consisted of annual averages from 2009–2010 as measured by the Interagency Monitoring for Protected Visual Environments (IMPROVE) and Chemical Speciation Network (CSN) of the U.S. Environmental Protection Agency (U.S. EPA 2009). The IMPROVE monitors are a nationwide network located mostly in remote areas. The CSN monitors are located in more urban areas. These two networks provide data that are evenly dispersed throughout the lower 48 states (Figure 1).

Figure 1 Locations of IMPROVE and CSN monitors and predicted national average PM2.5 component concentrations from final predictions models. (A) EC, (B) OC, (C) Si, and (D) S. Insets show predictions for St. Paul, MN.

All IMPROVE and CSN monitors that had at least 10 data points per quarter and a maximum of 45 days between measurements were included in our analyses. Si and S measurements were averaged over 1 January 2009–31 December 2009. The EC/OC data set consisted of measurements from 204 IMPROVE and CSN monitors averaged over 1 January 2009–31 December 2009, and measurements from 51 CSN monitors averaged over 1 May 2009–30 April 2010. We used the latter period because the measurement protocol used by CSN monitors prior to 1 May 2009 was incompatible with the IMPROVE network protocol. Comparing values averaged over 1 May 2009–30 April 2010 to those averaged over 1 January 2009–31 December 2009 indicated little difference between the time periods (data not shown). The annual averages were square-root transformed prior to modeling.

Geographic covariates. Approximately 600 LUR covariates were available for all monitor and subject locations. These included distances to A1, A2, and A3 roads [census feature class codes (CFCCs U.S. Census Bureau 2013)] land use within a given buffer population density within a given buffer and Normalized Difference Vegetation Index (NDVI National Oceanic and Atmospheric Administration 2013), which measures the level of vegetation in a monitor’s vicinity. CFCC A1 roads are limited-access highways A2 and A3 roads are other major roads such as county and state highways without limited access (Mercer et al. 2011). For NDVI a series of 23 monitor-specific, 16-day composite satellite images were obtained, and the pixels within a given buffer were averaged for each image. PLS incorporated the 25th, 50th, and 75th percentile of these 23 averages. The median of “high-vegetation season” image averages (defined as 1 April–30 September) and “low-vegetation season” averages (1 October–31 March) were also included. The geographic covariates were pre-processed to eliminate LUR covariates that were too homogeneous or outlier-prone to be of use. Specifically, we eliminated variables with > 85% identical values, and those with the most extreme standardized outlier > 7. We log-transformed and truncated all distance variables at 10 km, and computed additional “compiled” distance variables such as minimum distance to major roads and distance to any port. These compiled variables were then subject to the same inclusion criteria. All selected covariates were mean-centered and scaled by their respective SDs.

MESA cohort. MESA is a population-based study that began in 2000, with a cohort consisting of 6,814 participants from six U.S. cities: Los Angeles, California St. Paul, Minnesota Chicago, Illinois Winston-Salem, North Carolina New York, New York and Baltimore, Maryland. Four ethnic/racial groups were targeted: white, Chinese American, African American, and Hispanic. All participants were free of physician-diagnosed cardiovascular disease at time of entrance. [For additional details about the MESA study, see Bild et al. (2002).] These participants were also utilized in the Multi-Ethnic Study of Atherosclerosis and Air Pollution (MESA Air), an ancillary study to MESA funded by the U.S. EPA to study the relationship between chronic exposure to air pollution and progression of subclinical cardiovascular disease (Kaufman et al. 2012). Both the MESA and MESA Air studies were approved by the institutional review board (IRB) at each site, including the IRBs at the University of California, Los Angeles (Los Angeles, CA), Columbia University (New York, NY), Johns Hopkins University (Baltimore, MD), the University of Minnesota (Minneapolis-St. Paul, MN), Wake Forest University (Winston-Salem, NC), and Northwestern University (Evanston, IL). All subjects gave written informed consent.

We selected the CIMT end point in MESA as the health outcome for our case study. CIMT, a subclinical measure of atherosclerosis, was measured by B-mode ultrasound using a GE Logiq scanner (GE Healthcare, Wauwatosa, WI), and the end point was quantified as the right far wall CIMT measures conducted during MESA exam 1, which took place during 2000–2002 (Vedal et al., in press). We considered the 5,501 MESA participants who had CIMT measures during exam 1 our analysis was based on the 5,298 MESA participants who had CIMT measures during exam 1 and complete data for all selected model covariates.

Methods

The first stage of the two-stage approach included building the exposure models using PLS as the covariates in universal kriging models. We used cross-validation (CV) to select the number of PLS scores, determine how reliable predictions from each exposure model were, and assess the extent to which spatial structure was present for each pollutant. The health modeling stage of the two-stage approach included the health models we fit and the measurement error correction methods we employed. [For more detailed technical exposition, see Bergen et al. (2012).]

Spatial prediction models. Notation. Let Xt* denote the N* × 1 vector of observed square-root transformed concentrations at monitor locations R* the N* × p matrix of geographic covariates at monitor locations Xt the N × 1 vector of unknown square-root transformed concentrations at the unobserved subject locations and R the N × p matrix of geographic covariates at the subject locations. Note that for our exposure models, Xt* and Xt are dependent variables, and R* and R are independent variables. We used PLS to decompose R* into a set of linear combinations of much smaller dimension than R*. Specifically,

Here, H is a p × k matrix of weights for the geographic covariates, and T* is an N* × k matrix of PLS components or scores. These scores are linear combinations of the geographic covariates found in such a way that they maximize the covariance between Xt* and all possible linear combinations of R*. One might notice similarities between PLS and principal components analysis (PCA). Although the two methods are similar in that they are both dimension reduction methods, the scores from PLS maximize the covariance between Xt* and all other possible linear combinations of R*, whereas the scores from PCA are chosen to explain as much as possible the covariance of R*. [For more details see Sampson et al. (2013)]. PLS scores at unobserved locations are then derived as T = RH.

Once the PLS scores T and T* were obtained for the subject and monitoring locations, respectively, we assumed the following joint model for unobserved and observed exposures:

Here α is a vector of regression coefficients for the PLS scores, and η and η* are N × 1 and N* × 1 vectors of errors, respectively. Our primary exposure models assumed that the error terms exhibited spatial correlation that could be modeled with a kriging variogram parameterized by a vector of parameters θ⊇= (τ 2 , σ 2 , ϕ) (Cressie 1992). The nugget, τ 2 , is interpretable as the amount of variability in the pollution exposures that is not explained by spatial structure the partial sill, σ 2 , is interpretable as the amount of variability that is explained by spatial structure and the range, ϕ, is interpretable as the maximum distance between two locations beyond which they may no longer be considered spatially correlated. We estimated these parameters and the regression coefficients α via profile maximum likelihood. Once these parameters were estimated, we obtained predictions at unobserved locations by taking the mean of Xt conditional on Xt* and the estimated exposure model parameters. Because our measurement error correction methods rely on a correctly specified exposure model, we took care to choose the best-fitting kriging variogram to model our data. We initially fit exponential variograms for all four pollutants and investigated whether plots of the estimated variogram appeared to fit the empirical variogram well. If they appeared to fit poorly, we investigated spherical and cubic variograms. The exponential variogram fit well for EC, OC, and S, but provided a poor fit for Si (data not shown). We therefore examined cubic and spherical variograms and found the spherical variogram provided a much better fit and used it to model Si in our exposure models.

As a comparison to our primary kriging models, we also derived predictions from PLS alone without fitting a kriging variogram. This is analogous to a pure LUR model but using the PLS scores instead of actual geographic covariates. For this analysis η and η* were assumed to be independent, and α was estimated using a least-squares fit to regression of Xt* on T*. PLS-only predictions at the unobserved locations were then derived as the fitted values from this regression using the PLS scores at the subject locations.

CV and model selection. We used 10-fold CV (Hastie et al. 2001) to assess the models’ prediction accuracy, to select the number of PLS components to use in the final prediction models, and to compare predictions generated using PLS only to our primary models, which used both PLS and universal kriging. Data were randomly assigned to 1 of 10 groups. One group (a “test set”) was omitted, and the remaining groups (a “training set”) were used to fit the model and generate test set predictions. Each group played the role of test set until predictions were obtained for the entire data set. At each iteration, the following steps were taken to cross-validate our primary models (similar steps were followed to derive cross-validated predictions that used PLS only):

PLS was fit using the training set, and K scores were computed for the test set, for K = 1. 10.

Universal kriging parameters θ and coefficients α were estimated via profile maximum likelihood using the training set. The first K PLS scores correspond to T* in Equation 1, for K = 1. 10.

Predictions were derived using the first K PLS components and the corresponding universal kriging, using kriging parameters estimated from the training set.

We used the R package pls to fit the PLS. universal kriging was performed using the R package geoR. The best-performing models were selected out of those that used both PLS and kriging based on their cross-validated root mean squared error of prediction (RMSEP) and corresponding R 2 . For a data set with N* observations and corresponding predictions, the formulae for these performance metrics are given by

These metrics are sensitive to scale accordingly, they are useful for evaluating model performance for a given pollutant but not for comparing models across pollutants.

Health modeling. Disease model. Multivariable linear regression models were used to estimate the effects of each individual PM2.5 component exposure on CIMT. Each model included a single PM2.5 component along with a vector of subject-specific covariates. Let Y be the 5,298 × 1 vector of health outcomes for the 5,298 MESA participants included in the analysis, W the 5,298 × 1 vector of exposure predictions on the untransformed scale, and Z a matrix of potential confounders. We assumed linear relationships between Y, the true exposures, and Z, and fit the following equation via ordinary least squares (OLS):

E(Y) = β0 + Wβx + Zβz. [4]

Measurement error correction. The model in Equation 4 was fit using the predicted exposures W instead of the true exposures as the covariate of interest. Using predictions rather than true exposures in health modeling introduces two sources of measurement error that potentially influence the behavior of ^ βx. Berkson-like error arises from smoothing the true exposure surface and could inflate the SE of ^ βx. Classical-like error arises from estimating the exposure model parameters α and θ. The classical-like error potentially inflates the SE of ^ βx and could also bias the point estimate. We implemented the parameter bootstrap, an efficient method to assess and correct for the effects of measurement error. [See Szpiro et al. (2011b) for additional background and details.]

We used the parameter bootstrap in the context of predictions that use both PLS and universal kriging the approach would be very similar if PLS alone was used (although we did not implement that correction here).

Estimate a sampling density for ^ α and ^ θ with a multivariate normal distribution.

For j = 1. B bootstrap samples

Simulate new “observed” bootstrap exposures at monitoring locations from Equation 1 and health outcomes from Equation 4.

Sample new exposure model parameters and, from the sampling density estimated in step 1, using a constant covariance matrix multiplied by a scalar λ ≥ 0. λ controls the variability of ( ^ αj, ^ θj): the larger λ is, the greater the variability of ( ^ αj, ^ θj).

Use the simulated health outcomes and newly-sampled exposure model parameters to derive Wj.

Calculate ^ βx,j using Wj by OLS.

Let Eλ( ^ βx B ) denote the empirical mean of the ^ βx,j. The estimated bias is defined as Biasλ( ^ βx) = Eλ( ^ βx B )–E0( ^ βx B ) with corresponding bias-corrected effect estimate βx,λ corrected = ^ βx–Biasλ( ^ βx).

Estimate the bootstrap SE as

For our implementation of the parameter bootstrap, we set B = 30,000 and λ⊇= 1.

The goal of the parameter bootstrap is to approximate the sampling properties of the measurement error-impacted ^ βx that would be estimated if we performed our two-stage analysis with many actual realizations of monitoring observations and subject health data sets. Accordingly, step 2(a) gives us B new “realizations” of our data. For λ⊇= 1, step 2(b) accounts for the classical-like error by resampling the exposure model parameters. Step 2(c) accounts for the Berkson-like error by smoothing the true exposure surface. Step 2(d) then calculates B new ^ βx,j’s, the sampling properties of which have incorporated all sources of measurement error. Comparing these to the mean of bootstrapped ^ βx,j derived using fixed exposure model parameters (i.e., λ⊇= 0) gives us an approximation of the bias induced by the classical-like error (step 3), and the empirical SD approximates the SE that accounts for both sources of measurement error (step 4).

We also implemented the parameter bootstrap for λ = 0. This is equivalent to the “partial parametric bootstrap” described by Szpiro et al. (2011b), which accounts for the Berkson-like error only because the exposure surface is still smoothed, but with fixed parameters.

A desirable trait of the parameter bootstrap is the ability to “tune” the amount of the classical-like error by varying λ, which allows us to investigate how variability in the sampling distribution of ( ^ αj, ^ θj) affects the bias of ^ βx. This can be useful in refining our bootstrap bias estimates by simulation extrapolation (SIMEX) (Stefanski and Cook 1995). (For additional information on our approach to SIMEX and the results of applying it to the MESA data, see Supplemental Material, pp. 2–3 and Figure S1.)

Results

Data. Monitoring data. Mean concentrations of the four pollutants according to monitoring network are shown in Table 1. EC and OC concentrations measured by CSN monitors tended to be higher than concentrations measured by IMPROVE monitors. Average Si and S concentrations measured by CSN monitors were also higher than the IMPROVE averages however, relative to their SDs, the differences between CSN and IMPROVE monitors in Si and S concentrations were not as great as the differences between EC and OC concentrations.

Table 1 Summary data for observed pollution concentrations (mean ± SD) at monitoring networks predicted concentrations (mean ± SD) for the MESA cohort at exam 1 and summaries of selected LUR covariates.

CovariatesIMPROVECSNAll monitorsMESA Air
Sites (n)190982885501
EC (μg/m 3 )0.19±0.180.66±0.240.37±0.300.74±0.18
OC (μg/m 3 )0.93±0.552.23±0.711.43±0.882.17±0.36
Si (ng/m 3 )0.16±0.120.10±0.090.14±0.110.09±0.03
S (μg/m 3 )0.41±0.270.69±0.250.51±0.290.78±0.15
Sites <150m to an A1 road[n (%)]4 (2)3 (3)7 (2)249 (6)
Sites <150m to an A3 road[n (%)]36 (19)43 (44)79 (27)2,763 (50)
Median distance to comm (m)4,6961271,235302
Median pop dens a (persons/mi 2 )3805203,496
NDVI b150140146137
Abbreviations: comm, commercial or service centers pop dens, population density. a Persons per square mile for census block/block group to which monitor/­subject belongs. b Median value of summer NDVI medians within 250-m buffer.

Geographic covariates. The geographic variables that we used are listed in Table 2. Most of these variables were used for modeling all four pollutants, but not all. The following variables were used for modeling Si and S but not EC and OC: PM2.5 and PM10 emissions, streams and canals within a 3-km buffer, other urban or built-up land use within a 400-m buffer, lakes within a 10-km buffer, industrial and commercial complexes within a 15-km buffer, and herbaceous rangeland within a 3-km buffer. On the other hand, the following variables were used for modeling EC and OC but not Si and S: industrial land use within 1- and 1.5-km buffers.

Table 2 LUR covariates (Figure 2 abbreviations) and (where applicable) covariate buffer sizes that made it through preprocessing and were considered by PLS.

AbbreviationVariable descriptionBuffer sizes
Distance to features A1 road aNA
Airport aNA
Large airport aNA
Port aNA
Coastline a,bNA
Commercial or service center aNA
Rail yard aNA
SO2SO2 Emissions c30km
PM2.5PM2.5 c,d30km
PM10PM10 c,d30km
NOxNOx c30km
PopulationPopulation density500m, 1km, 1.5km, 2km, 2.5km, 3km, 5km, 10km, 15km
NDVI–winterMedian winter250m, 500m, 1km, 2.5km, 5km, 7.5km, 10km
NDVI–summerMedian summer250m, 500m, 1km, 2.5km, 5km, 7.5km, 10km
NDVI–Q7575th percentile250m, 500m, 1km, 2.5km, 5km, 7.5km, 10km
NDVI–Q5050th percentile250m, 500m, 1km, 2.5km, 5km, 7.5km, 10km
NDVI–Q2525th percentile250m, 500m, 1km, 2.5km, 5km, 7.5km, 10km
TransportTransportation, communities, and utilities750m, 3km, 5km, 10km, 15km
TransitionTransitional areas15km
StreamStreams and canals3km d , 5km, 10km, 15km
ShrubShrub and brush rangeland1.5km, 3km, 5km, 10km, 15km
ResidentialResidential400m, 500m, 750m, 1km, 1.5km, 3km, 5km, 10km, 15km
Other urbanOther urban or built-up400m d , 500m, 1.5km, 3km, 5km, 10km, 15km
Mixed rangeMixed rangeland3km, 5km, 10km, 15km
Mixed forestMixed forest land750m, 1km, 1.5km, 3km, 5km, 10km, 15km
LakesLakes d10 km
IndustrialIndustrial1km e , 1.5km e , 3km, 5km, 10km, 15km
Indust/commIndustrial and commercial complexes d15km
Herb rangeHerbaceous rangeland3km d , 5km, 10km
GreenEvergreen forest land400m, 500m, 750m, 1km, 1.5km, 3km, 5km, 10km, 15km
ForestDeciduous forest land750m, 1km, 1.5km, 3km, 5km, 10km, 15km
CropCropland and pasture400m, 500m, 750m, 1km, 1.5km, 3km, 5km, 10km, 15km
CommCommercial and services500m, 750m, 1km, 1.5km, 3km, 5km, 10km, 15km
A23Total distance of A2 and A3 roads within buffer100m, 150m, 300m, 400m, 500m, 750m, 1km, 1.5km, 3km, 5km
A1Total distance of A1 roads within buffer1km, 1.5km, 3km, 5km
Most variables were used in each of the four PM2.5 component models however, the pre-processing procedure selected some variables for EC and OC that were not selected for Si and S, and vice versa because EC and OC monitoring locations were not identical to Si and S locations. a Truncated at 25km and log10 transformed. b log10 and untransformed values both included. c Tons per year of emissions from tall stacks. d Variable used for modeling Si, S only. e Variable used for modeling EC and OC only.

The distributions of selected geographic covariates are shown according to monitoring network and MESA locations in Table 1. Although relatively few monitors belonging to either IMPROVE or CSN were within 150 m of an A1 road, there was a larger proportion of CSN monitors within 150 m of an A3 road (44%) than IMPROVE monitors (19%), consistent with the placement of CSN monitors in more urban locations compared with IMPROVE monitors (Table 1). The median distance to commercial and service centers was much smaller for CSN monitors (127 m vs. 4,696 m), and the median population density was much larger for CSN monitors (805 persons/mi 2 ) than for IMPROVE monitors (only 3 persons/mi 2 ). Median summer NDVI values within 250 m were slightly smaller for CSN monitors than for IMPROVE monitors, consistent with the placement of IMPROVE monitors in greener areas. Geographic covariate distributions among MESA participant locations were more consistent with the CSN monitors, as is especially evident for the number of sites < 150 m from an A3 road and median population density (Table 1). Density plots of the geographic covariates for monitoring and subject locations indicated noticeable overlap for all geographic covariates (data not shown), suggesting differences in geographic covariates between monitor and MESA locations were consistent with the concentration of MESA subjects in urban locations, not extrapolation beyond our data.

MESA cohort. Distributions of health model covariates among MESA cohort participants are summarized in Table 3. The mean CIMT (0.68 ± 0.19 mm) mean age (62 ± 10 years) sex (52% female) race (39% white, 12% Chinese American, 27% African American, and 22% Hispanic) and status (44% hypertension status and 15% statin use) were determined by questionnaire (Bild et al. 2002). The highest percentage of participants resided in Los Angeles (19.7%), but the distribution across the six cities was quite homogeneous. Only the 5,298 participants with complete data for all the selected model covariates listed in Table 3 were included in the analysis.

Table 3 Subject-specific covariates for the MESA cohort used in health modeling.

VariablenMean±SD or %
CIMT5,5010.68±0.19
Age (years)5,50161.9±10.1
Weight (lb)5,501173.0±37.5
Height (cm)5,501166.6±10.0
Waist (cm)5,50097.8±14.1
Body surface area (m 2 )5,5011.9±0.2
BMI (kg/m 2 )5,50128.2±5.3
DBP5,49971.8±10.3
Sex
Female2,87252.2
Male2,62947.8
Race
White (Caucasian)2,16839.4
Chinese American67512.3
Black (African American)1,45926.5
Hispanic1,19921.8
Site
Winston-Salem87816.0
New York86715.8
Baltimore77614.1
St. Paul and Minneapolis89916.3
Chicago99818.1
Los Angeles1,08319.7
Education
Incomplete high school91616.7
Completed high school99118.0
Some college1,57128.6
Completed college2,01036.5
Missing130.2
Income per year
<$12,00056610.3$12,000–24,9991,02218.6
$25,000–49,9991,54328$50,000–74,99990116.4
>$75,0001,27123.1 Missing1983.6 Hypertension No3,10656.5 Yes2,39543.5 Statin use No4,68185.1 Yes81714.9 Missing30.1 Spatial prediction models. Model evaluation. The selected models corresponding to lowest cross-validated R 2 all used PLS and universal kriging. For all four PM2.5 components and for all numbers of PLS scores, kriging improved prediction accuracy, as indicated by the R 2 and RMSEP statistics for the selected prediction models corresponding to the best performing PLS-only and PLS + universal kriging models (Table 4). Comparing the R 2 with and without universal kriging indicates that EC and OC were not much improved by kriging, whereas universal kriging improved prediction accuracy for Si and even more so for S. The ratio of the nugget to the sill (i.e., τ 2 /σ 2 ) also supports improved predictions with spatial smoothing by kriging. For a fixed range, smaller values of this ratio indicate that concentrations at nearby locations receive greater weight when kriging. We see this relationship in Table 4 where τ 2 /σ 2 was large when universal kriging did little to improve prediction accuracy, and very small when universal kriging helped improve prediction accuracy. Table 4 Cross-validated R 2 and RMSEP for each component of PM2.5, for both primary models and comparison PLS-only models, and the estimated kriging parame­ters from the likelihood fit on the entire data set for each pollutant. CorrectionModelECOCSiS 3 PLS scores2 PLS scores2 PLS scores2 PLS scores R 2 PLS only0.790.600.360.63 PLS+UK0.820.690.620.95 RMSEPPLS only0.110.220.100.13 PLS+UK0.100.200.080.05 Estimated UK parameters(τ 2 ) a0.00740.02510.00430.0007 (σ 2 ) b0.00250.01990.00860.0251 (φ) c4133042,7892,145 (τ 2 /σ 2 )2.961.260.50.03 UK, universal kriging. a Nugget used in kriging. b Partial sill used in kriging. c Range used in kriging. As a sensitivity analysis we also carried out CV using nearest-monitor exposure estimates. This method performed very poorly for EC and OC (R 2 s of 0 and 0.06, respectively), relatively poorly for Si (R 2 = 0.36), but performed well for S (R 2 = 0.88). Interpretation of PLS.Figure 2 illustrates the geographic covariates that were most important for explaining pollutant variability. Specifically, Figure 2 summarizes the p × 1 vector m, the vector such that Rm equals the 5,298 exposures predicted with PLS only. Each element of m is a weight for a corresponding geographic covariate. Positive elements in m (i.e., values > 0 in Figure 2) indicate that higher values of the geographic covariate were associated with higher predicted exposure the larger the absolute value of an element in m, the more the corresponding geographic covariate contributed to exposure prediction. Figure 2 Coefficients of the PLS fit, where the coefficients describe the associations of each geographic covariate with exposure for (A) EC, (B) OC, (C) Si, and (D) S. The size of each circle represents covariate buffer size, with larger circles indicating larger buffers. Each closed circle for “distance to feature” represents a different feature (listed in Table 2): A1 road, nearest road, airport, large airport, port, coastline, commercial or service center, railroad, and rail yard. Variable abbreviations and buffer sizes are indicated in Table 2. Most of the variables shown here were used for modeling all four pollutants, but not all. Variables used for modeling Si and S but not EC and OC were PM2.5 and PM10 emissions, streams and canals within a 3-km buffer, other urban or built-up land use within a 400-m buffer, lakes within a 10-km buffer, industrial and commercial complexes within a 15-km buffer, and herbaceous rangeland within a 3-km buffer. The variables used for modeling EC and OC but not Si and S were industrial land use within 1- and 1.5-km buffers. Population density was associated with larger predicted values of all pollutants, particularly for EC, OC, and S. Industrial land use within the smallest buffer was very predictive of EC and OC, and evergreen forest land within a given buffer was strongly predictive of decreases in S. NDVI, industrial land use, emissions, and line-length variables were positively associated with all exposures except Si, whereas all the distance-to-features variables were negatively associated with all exposures except Si. The NDVI variables were more important for prediction of OC and S than they were for EC. For Si, the NDVI and transitional land use variables appeared to be the most informative for prediction, with NDVI negatively and transitional land use positively associated with Si exposure. Distance to features appeared to be informative for all four pollutants. Exposure predictions. Figure 1 shows predicted concentrations across the United States, with finer detail illustrated for St. Paul, Minnesota. The EC and OC predictions were much higher in the middle of urban areas, and quickly dissipated further from urban centers. S predictions were high across the midwestern and eastern states and in the Los Angeles area, and lower in the plains and mountains. Si predictions were low in most urban areas, and high in desert states. Mean predicted EC and OC exposure concentrations predicted for MESA participants were 0.74 ± 0.18 and 2.17 ± 0.36 μg/m 3 , respectively (Table 1). Mean predicted Si and S exposure concentrations were 0.09 ± 0.03 ng/m 3 and 0.78 ± 0.15 μg/m 3 , respectively. Health models. The results from the naïve health model that did not include any measurement error correction, as well as the results from the health model that included bootstrap-corrected point estimates and SEs of ^ βx, are displayed in Table 5. The naïve analysis indicated significant positive associations (p < 0.05) of CIMT with OC, Si, and S. There was also a positive but nonsignificant association between CIMT and EC. SEs for the EC and OC health effects were virtually unchanged when measurement error correction was implemented, whereas the bootstrap-corrected SEs for Si and S were about 50% larger than their respective naïve estimates. The estimated biases resulting from the classical-like measurement error were so small as to be uninteresting from an epidemiologic perspective because the point estimates of all four pollutants after implementing measurement error correction were unchanged out to three decimal places. Table 5 Point estimates ± SEs and 95% CIs for the different pollutants, using naïve analysis and with bootstrap correction for measurement error in covariate of interest. PM2.5 componentAnalysis/­correctionβ̂x a ±SE95% CI EC (μg/m 3 )Naïve0.001±0.014–0.03, 0.03 PB, b λ=00.001±0.015–0.03, 0.03 PB, λ=10.001±0.015–0.03, 0.03 OC (μg/m 3 )Naïve0.025±0.0080.01, 0.04 PB, λ=00.025±0.0080.01, 0.04 PB, λ=10.025±0.0080.01, 0.04 Si (ng/m 3 )Naïve0.408±0.0810.25, 0.57 PB, λ=00.408±0.1260.16, 0.66 PB, λ=10.408±0.1270.16, 0.66 S (μg/m 3 )Naïve0.055±0.0170.022, 0.088 PB, λ=00.055±0.0250.006, 0.104 PB, λ=10.055±0.0250.006, 0.104 Point estimates are estimates of the increase in CIMT for a 1-unit increase in each pollutant. a In the case of λ=1, β̂x refers to the estimate corrected for any bias from ­classical-like error. b PB refers to results from parame­ter bootstrap implemented with given value of λ. Discussion Summary. Our comprehensive two-stage approach to estimating long-term effects of air pollution exposure includes a national prediction model to estimate exposures to individual PM2.5 components and corrects for measurement error in the epidemiologic analysis using a methodology that accounts for differing amounts of spatial structure in the exposure surfaces. In a case study of four components of PM2.5 and measurement error–corrected associations between these components and CIMT in the MESA cohort, corrected SEs corresponding to pollutants that exhibited significant spatial structure (i.e., Si and S) were 50% larger than naïve estimates, whereas corrected SE estimates for EC and OC were very similar to the naïve estimates. National exposure models. We find that a national approach to exposure modeling is reasonable and performs well in terms of prediction accuracy. Our primary PLS + universal kriging models resulted in cross-validated R 2 ≤ 0.95 (for predicting S concentrations) and ≥ 0.62 (for predicting Si) for any of the PM2.5 components. Use of kriging improved the cross-validated R 2 for all four pollutants compared with models that used PLS only, although the improvement was not equal across all four pollutants. These results are useful in terms of understanding the spatial nature of our exposure surfaces. For EC and OC, the R 2 only improved by ≤ 0.09 when kriging was used compared to when PLS alone was used, indicating little large-scale spatial structure in these pollutants. For Si, the R 2 improved from 0.36 to 0.62 and for S, from 0.63 to 0.95. This indicates that S (and to a lesser extent Si) had substantial large-scale spatial structure that kriging was able to exploit. For all models, using kriging improved R 2 , indicating that no prediction accuracy was lost (and quite a bit stood to be gained, when spatial structure was present) by using PLS+universal kriging as opposed to using PLS alone. Our results also suggest that exposure models such as the ones we have built may be preferable in many cases to simpler approaches such as nearest-monitor interpolation. Our models produced cross-validated R 2 that were higher than the nearest-monitor approach, and our results indicate that unless there is considerable spatial structure in the exposure surface, a substantial amount of prediction accuracy may be lost when the nearest-monitor approach is used. We used two-stage modeling instead of joint modeling of exposure and health for a variety of reasons. One is pragmatic: Joint modeling is computationally intensive, so our two-stage approach is especially desirable when modeling multiple pollutants. Joint modeling may also be more sensitive to outliers in the health data. Two-stage modeling also appeals more intuitively in the context of modeling multiple health outcomes because it assigns one exposure per participant that can then be used to model a number of different health outcomes. Joint modeling, on the other hand, would assign different levels of the same pollutant depending on what health outcome was being modeled. Epidemiologic case study. In this case study, we focused on four PM2.5 components selected to gain insight into the sources or features of PM2.5 that might contribute to the effects of PM2.5 on cardiovascular disease. EC and OC were chosen as markers of primary emissions from combustion processes, with OC also including contributions from secondary organic aerosols formed from atmospheric chemical reactions Si was chosen as a marker of crustal dust and S was chosen as a marker of sulfate, an inorganic aerosol formed secondarily from atmospheric chemical reactions (Vedal et al., in press). The mechanisms whereby exposures to PM2.5 or PM2.5 components produce cardiovascular effects such as atherosclerosis are not well understood, although several mechanisms have been proposed (Brook et al. 2010). [For discussion of other studies examining the effects of these pollutants, see Vedal et al. (in press).] The relatively poor performance of nearest-monitor interpolation for EC, OC, and Si raises concerns about epidemiologic inferences based on predictions derived from that method. For S, the only pollutant for which our models and nearest-monitor interpolation performed comparably, the estimated increase in CIMT for a 1-unit increase in exposure based on nearest-monitor interpolation was 0.074 ± 0.018, comparable to the naïve inference made using predictions from our exposure models (0.055 ± 0.017). However, there is no way to correct for measurement error using this method, which is another significant advantage of our models. Naïve health analyses based on exposure predictions from our national models indicated significant associations of CIMT with 1-unit increases in average OC, Si, and S, but not EC. Using the parameter bootstrap to account and correct for measurement error led to noticeably larger SEs and wider CIs for Si and S however, OC, Si, and S were still significantly associated with CIMT even after correcting for measurement error. Measurement error correction. For EC and OC, using PLS alone was sufficient to make accurate predictions, whereas the spatial smoothing from universal kriging substantially improved prediction accuracy for Si and S. It is accordingly no coincidence that the bootstrap-corrected SE estimates for EC and OC were unchanged from the naïve estimates, whereas the corrected SE estimates for Si and S were about 50% larger (and the resulting 95% CIs 50% wider) than their respective naïve estimates. The fact that the EC and OC exposure predictions were derived mostly from the PLS-only models, which assumed independent residuals, implies that the Berkson-like error was almost pure Berkson error (i.e., independent across location), which was correctly accounted for by naïve SE estimates. On the other hand, much more smoothing took place for Si and S, which induced spatial correlation in the residual difference between true and predicted exposure. Accordingly, SEs that correctly account for the Berkson-like error in these two pollutants are inflated because the correlated errors in the predictions translate into correlated residuals in the disease model that are not accounted for by naïve SE estimates (Szpiro et al. 2011b). The fact that the SE estimates from the parameter bootstrap using λ = 1 (which accounts for both Berkson-like and classical-like error) and using λ = 0 (which accounts only for Berkson-like error) were so similar further indicates that the larger corrected SE estimates were most likely a result of the Berkson-like error. None of our measurement error analyses indicated that any important bias was induced by the classical-like error. Limitations and model considerations. Although our exposure models performed well, there is still room for improvement in prediction accuracy, especially for the EC, OC, and Si models, which had cross-validated R 2 that could be improved upon. For these models it is possible that inclusion of additional geographic covariates in the PLS would help improve model performance. Examples include wood-burning sources within a given buffer for EC and OC concentrations, or dust and sand sources for Si. These covariates are currently not available in our databases. Furthermore, although it is possible to interpret the individual covariates in PLS components (Figure 2), such interpretations need to be regarded with caution because inclusion of many correlated covariates can lead to apparent associations that are counter-intuitive and the opposite of what might be expected scientifically. Finally, PLS does not consider interactions or nonlinear combinations of the geographic covariates, factors which could improve model performance. Implications and future directions. Our results show that careful investigation of the exposure model characteristics can help to clarify the implications for the subsequent epidemiologic analyses that use the predicted exposures. As noted by Szpiro et al. (2011a), an overarching framework that considers the end goal of health modeling seems more appealing than treating exposure models as if they exist for their own sake. This analysis serves as an example that will inform ongoing efforts by our group and others to construct and utilize exposure prediction models that are most suitable for epidemiologic studies. Our epidemiologic inference was based on one health model per pollutant. One might reasonably be interested in how multiple pollutants jointly affect health. However, current literature for measurement error correction does not address models that use multiple predicted pollutants as exposures. Our group is currently working on methods to address this challenge. Supplemental Material (2.4 MB) PDF Click here for additional data file. We thank the three reviewers for their helpful comments. Research in this publication was supported by grants T32ES015459, P50ES015915, and R01ES009411 from the National Institute of Environmental Health Sciences of the National Institutes of Health (NIH). Additional support was provided by an award to the University of Washington under the National Particle Component Toxicity initiative of the Health Effects Institute and by the U.S. Environmental Protection Agency (EPA), Assistance Agreement RD-83479601-0 (Clean Air Research Centers). This publication was developed under a STAR (Science to Achieve Results) program research assistance agreement, RD831697, awarded by the U.S. EPA. The views expressed in this document are solely those of the University of Washington, and the U.S. EPA does not endorse any products or commercial services mentioned in this publication. The Multi-Ethnic Study of Atherosclerosis (MESA) is conducted and supported by the National Heart, Lung, and Blood Institute (NHLBI) in collaboration with MESA investigators. Support for MESA is provided by NHLBI contracts N01HC-95159 through N01HC95169 and UL1RR024156. MESA Air is funded by the U.S. EPA’s STAR grant RD831697. The content of this work is solely the responsibility of the authors and does not necessarily represent the official views of the NIH. The authors declare they have no actual or potential competing financial interests. Can't convert from a BYTE buffer to a cv::Mat image So i have a problem with converting a BYTE buffer to an image, (cv::Mat). I am trying to read a real time video from a distant camera, and i got two elements, a pointer to the buffer and the buffer size, and i need to convert that to a cv::Mat image so that i can show it with cv::imshow. i tried to use: but it isn't working and i get this error: when i try to convert directly without the imdecode function like this: i get an image but i can't show it so the programm just continue running without doing anything. Can anyone help me please on how do we convert from BYTE buffer pointer to an cv::Mat image Buffer is declared like this : BYTE *Buffer the function where i get the buffer from is declared like this where: lRealHandle : Real-time monitoring handle dwDataType : pBuffer : Buffer for callback data. Data of different length will be called back according to different data type. The data are called back by frame for every type but type 0, and each time one frame is called back. dwBufSize : Callback data length. The data buffers are diffreent for different types. The unit is BYTE In my case i always get data type 0 This is how i try to decode then : my programm stops here, it continue running but it dosen't do anything after cuz i have put a std::cout here to check if it will pass this line of imshow or not but nothing happens Contents The term "Middle East" may have originated in the 1850s in the British India Office. [6] However, it became more widely known when American naval strategist Alfred Thayer Mahan used the term in 1902 [7] to "designate the area between Arabia and India". [8] [9] During this time the British and Russian Empires were vying for influence in Central Asia, a rivalry which would become known as The Great Game. Mahan realized not only the strategic importance of the region, but also of its center, the Persian Gulf. [10] [11] He labeled the area surrounding the Persian Gulf as the Middle East, and said that after Egypt's Suez Canal, it was the most important passage for Britain to control in order to keep the Russians from advancing towards British India. [12] Mahan first used the term in his article "The Persian Gulf and International Relations", published in September 1902 in the National Review, a British journal. The Middle East, if I may adopt a term which I have not seen, will some day need its Malta, as well as its Gibraltar it does not follow that either will be in the Persian Gulf. Naval force has the quality of mobility which carries with it the privilege of temporary absences but it needs to find on every scene of operation established bases of refit, of supply, and in case of disaster, of security. The British Navy should have the facility to concentrate in force if occasion arise, about Aden, India, and the Persian Gulf. [13] Mahan's article was reprinted in The Times and followed in October by a 20-article series entitled "The Middle Eastern Question," written by Sir Ignatius Valentine Chirol. During this series, Sir Ignatius expanded the definition of Middle East to include "those regions of Asia which extend to the borders of India or command the approaches to India." [14] After the series ended in 1903, The Times removed quotation marks from subsequent uses of the term. [15] Until World War II, it was customary to refer to areas centered around Turkey and the eastern shore of the Mediterranean as the "Near East", while the "Far East" centered on China, [16] and the Middle East then meant the area from Mesopotamia to Burma, namely the area between the Near East and the Far East. [ citation needed ] In the late 1930s, the British established the Middle East Command, which was based in Cairo, for its military forces in the region. After that time, the term "Middle East" gained broader usage in Europe and the United States, with the Middle East Institute founded in Washington, D.C. in 1946, among other usage. [17] The corresponding adjective is Middle Eastern and the derived noun is Middle Easterner. While non-Eurocentric terms such "Southwest Asia" or "Swasia" has been sparsedly used, the inclusion of an African country, Egypt, in the definition questions the usefulness of using such terms. [18] Criticism and usage The description Middle has also led to some confusion over changing definitions. Before the First World War, "Near East" was used in English to refer to the Balkans and the Ottoman Empire, while "Middle East" referred to Iran, the Caucasus, Afghanistan, Central Asia, and Turkestan. In contrast, "Far East" referred to the countries of East Asia (e.g. China, Japan, Korea, etc.) With the disappearance of the Ottoman Empire in 1918, "Near East" largely fell out of common use in English, while "Middle East" came to be applied to the re-emerging countries of the Islamic world. However, the usage "Near East" was retained by a variety of academic disciplines, including archaeology and ancient history, where it describes an area identical to the term Middle East, which is not used by these disciplines (see Ancient Near East). The first official use of the term "Middle East" by the United States government was in the 1957 Eisenhower Doctrine, which pertained to the Suez Crisis. Secretary of State John Foster Dulles defined the Middle East as "the area lying between and including Libya on the west and Pakistan on the east, Syria and Iraq on the North and the Arabian peninsula to the south, plus the Sudan and Ethiopia." [16] In 1958, the State Department explained that the terms "Near East" and "Middle East" were interchangeable, and defined the region as including only Egypt, Syria, Israel, Lebanon, Jordan, Iraq, Saudi Arabia, Kuwait, Bahrain, and Qatar. [19] The Associated Press Stylebook says that Near East formerly referred to the farther west countries while Middle East referred to the eastern ones, but that now they are synonymous. It instructs: Use Middle East unless Near East is used by a source in a story. Mideast is also acceptable, but Middle East is preferred. [20] The term Middle East has also been criticised as Eurocentric ("based on a British Western perception") by Hanafi (1998). [21] Translations There are terms similar to Near East and Middle East in other European languages, but since it is a relative description, the meanings depend on the country and are different from the English terms generally. In German the term Naher Osten (Near East) is still in common use (nowadays the term Mittlerer Osten is more and more common in press texts translated from English sources, albeit having a distinct meaning) and in Russian Ближний Восток or Blizhniy Vostok, Bulgarian Близкия Изток, Polish Bliski Wschód or Croatian Bliski istok (meaning Near East in all the four Slavic languages) remains as the only appropriate term for the region. However, some languages do have "Middle East" equivalents, such as the French Moyen-Orient, Swedish Mellanöstern, Spanish Oriente Medio or Medio Oriente, and the Italian Medio Oriente. [note 1] Perhaps because of the influence of the Western press, the Arabic equivalent of Middle East (Arabic: الشرق الأوسط ash-Sharq al-Awsaṭ) has become standard usage in the mainstream Arabic press, comprising the same meaning as the term "Middle East" in North American and Western European usage. The designation, Mashriq, also from the Arabic root for East, also denotes a variously defined region around the Levant, the eastern part of the Arabic-speaking world (as opposed to the Maghreb, the western part). [22] Even though the term originated in the West, apart from Arabic, other languages of countries of the Middle East also use a translation of it. The Persian equivalent for Middle East is خاورمیانه (Khāvar-e miyāneh), the Hebrew is המזרח התיכון (hamizrach hatikhon) and the Turkish is Orta Doğu. Territories and regions usually considered within the Middle East Traditionally included within the Middle East are Iran (Persia), Asia Minor, Mesopotamia, the Levant, the Arabian Peninsula, and Egypt. In modern-day-country terms they are these: Arms Flag State Area (km 2 ) Population (2012) [ needs update ] Density (per km 2 ) Capital Nominal GDP, bn (2018) [23] Per capita (2018) [24] Currency Government Official languages Akrotiri and Dhekelia 254 15,700 N/A Episkopi N/A N/A Euro De facto stratocratic dependency under a constitutional monarchy English Bahrain 780 1,234,596 1,582.8 Manama$30.355 $25,851 Bahraini dinar Absolute monarchy Arabic Cyprus 9,250 1,088,503 117 Nicosia$24.492 $28,340 Euro Presidential republic Greek, Turkish Egypt 1,010,407 82,798,000 90 Cairo$249.559 $2,573 Egyptian pound Presidential republic Arabic Iran 1,648,195 78,868,711 45 Tehran$452.275 $5,491 Iranian rial Islamic republic Persian Iraq 438,317 33,635,000 73.5 Baghdad$226.07 $5,930 Iraqi dinar Parliamentary republic Arabic, Kurdish Israel 20,770 7,653,600 365.3 Jerusalem a$369.843 $41,644 Israeli shekel Parliamentary republic Hebrew Jordan 92,300 6,318,677 68.4 Amman$42.371 $4,278 Jordanian dinar Constitutional monarchy Arabic Kuwait 17,820 3,566,437 167.5 Kuwait City$141.05 $30,839 Kuwaiti dinar Constitutional monarchy Arabic Lebanon 10,452 4,228,000 404 Beirut$56.409 $9,257 Lebanese pound Parliamentary republic Arabic Oman 212,460 2,694,094 9.2 Muscat$82.243 $19,302 Omani rial Absolute monarchy Arabic Palestine 6,220 4,260,636 667 Ramallah a n/a n/a Israeli shekel, Jordanian dinar Semi-presidential republic Arabic Qatar 11,437 1,696,563 123.2 Doha$192.45 $70,780 Qatari riyal Absolute monarchy Arabic Saudi Arabia 2,149,690 27,136,977 12 Riyadh$782.483 $23,566 Saudi riyal Absolute monarchy Arabic Syria 185,180 23,695,000 118.3 Damascus n/a n/a Syrian pound Presidential republic Arabic Turkey 783,562 73,722,988 94.1 Ankara$766.428 $9,346 Turkish lira Presidential republic Turkish United Arab Emirates 82,880 8,264,070 97 Abu Dhabi$424.635 $40,711 UAE dirham Federal Absolute monarchy Arabic Yemen 527,970 23,580,000 44.7 Sana'a b Aden (provisional)$26.914 $872 Yemeni rial Provisional presidential republic Arabic a. ^ ^ Jerusalem is the proclaimed capital of Israel, which is disputed and the actual location of the Knesset, Israeli Supreme Court, and other governmental institutions of Israel. Ramallah is the actual location of the government of Palestine, whereas the proclaimed capital of Palestine is East Jerusalem, which is disputed. b. ^ Controlled by the Houthis due to the ongoing war. Seat of government moved to Aden. Other definitions of the Middle East Various concepts are often being paralleled to Middle East, most notably Near East, Fertile Crescent and the Levant. Near East, Levant and Fertile Crescent are geographic concepts, which refer to large sections of the modern defined Middle East, with Near East being the closest to Middle East in its geographic meaning. Due to it primarily being Arabic speaking, the Maghreb region of North Africa is sometimes included. The countries of the South Caucasus—Armenia, Azerbaijan, and Georgia—are occasionally included in definitions of the Middle East. [25] The Greater Middle East was a political term coined by the second Bush administration in the first decade of the 21st century, [26] to denote various countries, pertaining to the Muslim world, specifically Iran, Turkey, Afghanistan and Pakistan. [27] Various Central Asian countries are sometimes also included. [28] The Middle East lies at the juncture of Eurasia and Africa and of the Mediterranean Sea and the Indian Ocean. It is the birthplace and spiritual center of religions such as Christianity, Islam, Judaism, Manichaeism, Yezidi, Druze, Yarsan and Mandeanism, and in Iran, Mithraism, Zoroastrianism, Manicheanism, and the Baháʼí Faith. Throughout its history the Middle East has been a major center of world affairs a strategically, economically, politically, culturally, and religiously sensitive area. The region is one of the regions were agriculture was independently discovered, and from the Middle East it was spread, during the Neolithic, to different regions of the world such as Europe, the Indus Valley and Eastern Africa. Prior to the formation of civilizations, advanced cultures formed all over the Middle East during the Stone Age. The search for agricultural lands by agriculturalists, and pastoral lands by herdsmen meant different migrations took place within the region and shaped its ethnic and demographic makeup. The Middle East is widely and most famously known as the Cradle of civilization. The world's earliest civilizations, Mesopotamia (Sumer, Akkad, Assyria and Babylonia), ancient Egypt and Kish in the Levant, all originated in the Fertile Crescent and Nile Valley regions of the ancient Near East. These were followed by the Hittite, Greek, Hurrian and Urartian civilisations of Asia Minor Elam, Persia and Median civilizations in Iran, as well as the civilizations of the Levant (such as Ebla, Mari, Nagar, Ugarit, Canaan, Aramea, Mitanni, Phoenicia and Israel) and the Arabian Peninsula (Magan, Sheba, Ubar). The Near East was first largely unified under the Neo Assyrian Empire, then the Achaemenid Empire followed later by the Macedonian Empire and after this to some degree by the Iranian empires (namely the Parthian and Sassanid Empires), the Roman Empire and Byzantine Empire. The region served as the intellectual and economic center of the Roman Empire and played an exceptionally important role due to its periphery on the Sassanid Empire. Thus, the Romans stationed up to five or six of their legions in the region for the sole purpose of defending it from Sassanid and Bedouin raids and invasions. From the 4th century CE onwards, the Middle East became the center of the two main powers at the time, the Byzantine empire and the Sassanid Empire. However, it would be the later Islamic Caliphates of the Middle Ages, or Islamic Golden Age which began with the Islamic conquest of the region in the 7th century AD, that would first unify the entire Middle East as a distinct region and create the dominant Islamic Arab ethnic identity that largely (but not exclusively) persists today. The 4 caliphates that dominated the Middle East for more than 600 years were the Rashidun Caliphate, the Umayyad caliphate, the Abbasid caliphate and the Fatimid caliphate. Additionally, the Mongols would come to dominate the region, the Kingdom of Armenia would incorporate parts of the region to their domain, the Seljuks would rule the region and spread Turko-Persian culture, and the Franks would found the Crusader states that would stand for roughly two centuries. Josiah Russell estimates the population of what he calls "Islamic territory" as roughly 12.5 million in 1000 – Anatolia 8 million, Syria 2 million, and Egypt 1.5 million. [29] From the 16th century onward, the Middle East came to be dominated, once again, by two main powers: the Ottoman Empire and the Safavid dynasty. The modern Middle East began after World War I, when the Ottoman Empire, which was allied with the Central Powers, was defeated by the British Empire and their allies and partitioned into a number of separate nations, initially under British and French Mandates. Other defining events in this transformation included the establishment of Israel in 1948 and the eventual departure of European powers, notably Britain and France by the end of the 1960s. They were supplanted in some part by the rising influence of the United States from the 1970s onwards. In the 20th century, the region's significant stocks of crude oil gave it new strategic and economic importance. Mass production of oil began around 1945, with Saudi Arabia, Iran, Kuwait, Iraq, and the United Arab Emirates having large quantities of oil. [30] Estimated oil reserves, especially in Saudi Arabia and Iran, are some of the highest in the world, and the international oil cartel OPEC is dominated by Middle Eastern countries. During the Cold War, the Middle East was a theater of ideological struggle between the two superpowers and their allies: NATO and the United States on one side, and the Soviet Union and Warsaw Pact on the other, as they competed to influence regional allies. Besides the political reasons there was also the "ideological conflict" between the two systems. Moreover, as Louise Fawcett argues, among many important areas of contention, or perhaps more accurately of anxiety, were, first, the desires of the superpowers to gain strategic advantage in the region, second, the fact that the region contained some two-thirds of the world's oil reserves in a context where oil was becoming increasingly vital to the economy of the Western world [. ] [31] Within this contextual framework, the United States sought to divert the Arab world from Soviet influence. Throughout the 20th and 21st centuries, the region has experienced both periods of relative peace and tolerance and periods of conflict particularly between Sunnis and Shiites. Ethnic groups Arabs constitute the largest ethnic group in the Middle East, followed by various Iranian peoples and then by Turkic speaking groups (Turkish, Azeris, and Iraqi Turkmen). Native ethnic groups of the region include, in addition to Arabs, Arameans, Assyrians, Baloch, Berbers, Copts, Druze, Greek Cypriots, Jews, Kurds, Lurs, Mandaeans, Persians, Samaritans, Shabaks, Tats, and Zazas. European ethnic groups that form a diaspora in the region include Albanians, Bosniaks, Circassians (including Kabardians), Crimean Tatars, Greeks, Franco-Levantines, Italo-Levantines, and Iraqi Turkmens. Among other migrant populations are Chinese, Filipinos, Indians, Indonesians, Pakistanis, Pashtuns, Romani, and Afro-Arabs. Migration "Migration has always provided an important vent for labor market pressures in the Middle East. For the period between the 1970s and 1990s, the Arab states of the Persian Gulf in particular provided a rich source of employment for workers from Egypt, Yemen and the countries of the Levant, while Europe had attracted young workers from North African countries due both to proximity and the legacy of colonial ties between France and the majority of North African states." [32] According to the International Organization for Migration, there are 13 million first-generation migrants from Arab nations in the world, of which 5.8 reside in other Arab countries. Expatriates from Arab countries contribute to the circulation of financial and human capital in the region and thus significantly promote regional development. In 2009 Arab countries received a total of US$35.1 billion in remittance in-flows and remittances sent to Jordan, Egypt and Lebanon from other Arab countries are 40 to 190 per cent higher than trade revenues between these and other Arab countries. [33] In Somalia, the Somali Civil War has greatly increased the size of the Somali diaspora, as many of the best educated Somalis left for Middle Eastern countries as well as Europe and North America.

Non-Arab Middle Eastern countries such as Turkey, Israel and Iran are also subject to important migration dynamics.

A fair proportion of those migrating from Arab nations are from ethnic and religious minorities facing racial and or religious persecution and are not necessarily ethnic Arabs, Iranians or Turks. [ citation needed ] Large numbers of Kurds, Jews, Assyrians, Greeks and Armenians as well as many Mandeans have left nations such as Iraq, Iran, Syria and Turkey for these reasons during the last century. In Iran, many religious minorities such as Christians, Baháʼís and Zoroastrians have left since the Islamic Revolution of 1979. [ citation needed ]

Religions

The Middle East is very diverse when it comes to religions, many of which originated there. Islam is the largest religion in the Middle East, but other faiths that originated there, such as Judaism and Christianity, are also well represented. Christians represent 40.5% of Lebanon, where the Lebanese president, half of the cabinet, and half of the parliament follow one of the various Lebanese Christian rites. There are also important minority religions like the Baháʼí Faith, Yarsanism, Yazidism, Zoroastrianism, Mandaeism, Druze, and Shabakism, and in ancient times the region was home to Mesopotamian religions, Canaanite religions, Manichaeism, Mithraism and various monotheist gnostic sects.

Languages

The six top languages, in terms of numbers of speakers, are Arabic, Persian, Turkish, Kurdish, Hebrew and Greek. Arabic and Hebrew represent the Afro-Asiatic language family. Persian, Kurdish and Greek belong to the Indo-European language family. Turkish belongs to Turkic language family. About 20 minority languages are also spoken in the Middle East.

Arabic, with all its dialects, is the most widely spoken language in the Middle East, with Literary Arabic being official in all North African and in most West Asian countries. Arabic dialects are also spoken in some adjacent areas in neighbouring Middle Eastern non-Arab countries. It is a member of the Semitic branch of the Afro-Asiatic languages. Several Modern South Arabian languages such as Mehri and Soqotri are also spoken Yemen and Oman. Another Semitic language such as Aramaic and its dialects are spoken mainly by Assyrians and Mandaeans. There is also an Oasis Berber-speaking community in Egypt where the language is also known as Siwa. It is a non-Semitic Afro-Asiatic language.

Persian is the second most spoken language. While it is primarily spoken in Iran and some border areas in neighbouring countries, the country is one of the region's largest and most populous. It belongs to the Indo-Iranian branch of the family of Indo-European languages. Other Western Iranic languages spoken in the region include Achomi, Daylami, Kurdish dialects, Semmani, Lurish, amongst many others.

The third-most widely spoken language, Turkish, is largely confined to Turkey, which is also one of the region's largest and most populous countries, but it is present in areas in neighboring countries. It is a member of the Turkic languages, which have their origins in Central Asia. Another Turkic language, Azerbaijani, is spoken by Azerbaijanis in Iran.

Hebrew is one of the two official languages of Israel, the other being Arabic. Hebrew is spoken and used by over 80% of Israel's population, the other 20% using Arabic.

Greek is one of the two official languages of Cyprus, and the country's main language. Small communities of Greek speakers exist all around the Middle East until the 20th century it was also widely spoken in Asia Minor (being the second most spoken language there, after Turkish) and Egypt. During the antiquity, Ancient Greek was the lingua franca for many areas of the western Middle East and until the Muslim expansion it was widely spoken there as well. Until the late 11th century, it was also the main spoken language in Asia Minor after that it was gradually replaced by the Turkish language as the Anatolian Turks expanded and the local Greeks were assimilated, especially in the interior.

English is one of the official languages of Akrotiri and Dhekelia. [34] [35] It is also commonly taught and used as a second language, especially among the middle and upper classes, in countries such as Egypt, Jordan, Iran, Kurdistan, Iraq, Qatar, Bahrain, United Arab Emirates and Kuwait. [36] [37] It is also a main language in some Emirates of the United Arab Emirates.

French is taught and used in many government facilities and media in Lebanon, and is taught in some primary and secondary schools of Egypt and Syria. Maltese, a Semitic language mainly spoken in Europe, is also used by the Franco-Maltese diaspora in Egypt.

Armenian speakers are also to be found in the region. Georgian is spoken by the Georgian diaspora. Russian is spoken by a large portion of the Israeli population, because of emigration in the late 1990s. [38] Russian today is a popular unofficial language in use in Israel news, radio and sign boards can be found in Russian around the country after Hebrew and Arabic. Circassian is also spoken by the diaspora in the region and by almost all Circassians in Israel who speak Hebrew and English as well. The largest Romanian-speaking community in the Middle East is found in Israel, where as of 1995 [update] Romanian is spoken by 5% of the population. [note 2] [39] [40]

Bengali, Hindi and Urdu are widely spoken by migrant communities in many Middle Eastern countries, such as Saudi Arabia (where 20–25% of the population is South Asian), the United Arab Emirates (where 50–55% of the population is South Asian), and Qatar, which have large numbers of Pakistani, Bangladeshi and Indian immigrants.

How to find &ldquoempty&rdquo buffer? - Geographic Information Systems

Academia.edu no longer supports Internet Explorer.

Stu Hamilton is an Associate Professor of Geography and Geosciences at Salisbury University as well as being the GIS Graduate Program Director. His research interests are global environmental change and human-environment interaction. He is a Fellow of the Royal Society of Geographers (FRGS) as well as a certified GIS Professional (GISP). He holds a BS in Geography and Applied Social Science from Canterbury Christ Church University in the UK, a MA in Geography (GIS) from the SUNY at Buffalo, and a PhD in Geography from the University of Southern Mississippi.

Aim To provide high-resolution local, regional, national and global estimates of annual mangrove. more Aim
To provide high-resolution local, regional, national and global estimates of annual mangrove forest area from 2000 through to 2012 with the goal of driving mangrove research questions pertaining to biodiversity, carbon stocks, climate change, functionality, food security, livelihoods, fisheries support and conservation that have been impeded until now by a lack of suitable data.

Location
Global, covering 99% of all mangrove forests.

Methods
We synthesized the Global Forest Change database, the Terrestrial Ecosystems of the World database and the Mangrove Forests of the World database to extract mangrove forest cover at high spatial and temporal resolutions. We then used the new database to monitor mangrove cover at the global, national and protected area scales.

Results
Countries showing relatively high amounts of mangrove loss include Myanmar, Malaysia, Cambodia, Indonesia and Guatemala. Indonesia remains by far the largest mangrove-holding nation, containing between 26% and 29% of the global mangrove inventory with a deforestation rate of between 0.26% and 0.66% per year. We have made our new database, CGMFC-21, freely available.

Main conclusions
Global mangrove deforestation continues but at a much reduced rate of between 0.16% and 0.39% per year. Southeast Asia is a region of concern with mangrove deforestation rates between 3.58% and 8.08%, this in a region containing half of the entire global mangrove forest inventory. The global mangrove deforestation pattern from 2000 to 2012 is one of decreasing rates of deforestation, with many nations essentially stable, with the exception of the largest mangrove-holding region of Southeast Asia. We provide a standardized spatial dataset that monitors mangrove deforestation globally at high spatio-temporal resolutions. These data can be used to drive the mangrove research agenda, particularly as it pertains to monitoring of mangrove carbon stocks and the establishment of baseline local mangrove forest inventories required for payment for ecosystem service initiatives.

750, 1450 and 4550 m) that are statistically significant at the 95% confidence level. Cospectral analysis suggests that the variation in dune morphology is correlated with transverse ridges on the inner-shelf, the backbarrier cuspate headlands, and the historical and storm-related trends in shoreline change. Sections of the coast with little to no dune development before Hurricane Ivan were observed in the narrowest portions of the island (between headlands), west of the transverse ridges. Overwash penetration tended to be larger in these areas and island breaching was common, leaving the surface close to the watertable and covered by a lag of shell and gravel. In contrast, large foredunes and the backbarrier dunes were observed at the widest sections of the island (the cuspate headlands) and at crest of the transverse ridges. Due to the large dunes and the presence of the backbarrier dunes, these areas experienced less overwash penetration and most of the sediment from the beachface and dunes was deposited within the upper-shoreface. It is argued that this sediment is returned to the beachface through nearshore bar migration following the storm and that the areas with larger foredunes and backbarrier dunes have smaller rates of historical shoreline erosion compared to areas with smaller dunes and greater transfer of sediment to the washover terrace. Since the recovery of the dunes will vary depending on the availability of sediment from the washover and beachface, it is further argued that the alongshore pattern of dune morphology and the response of the island to the next extreme storm is forced by the transverse ridges and island width through alongshore variations in storm surge and overwash gradients respectively. These findings may be particularly important for coastal managers involved in the repair and rebuilding of coastal infrastructure that was damaged or destroyed during Hurricane Ivan.

750, 1450 and 4550 m) that are statistically significant at the 95% confidence level. Cospectral analysis suggests that the variation in dune morphology is correlated with transverse ridges on the inner-shelf, the backbarrier cuspate headlands, and the historical and storm-related trends in shoreline change. Sections of the coast with little to no dune development before Hurricane Ivan were observed in the narrowest portions of the island (between headlands), west of the transverse ridges. Overwash penetration tended to be larger in these areas and island breaching was common, leaving the surface close to the watertable and covered by a lag of shell and gravel. In contrast, large foredunes and the backbarrier dunes were observed at the widest sections of the island (the cuspate headlands) and at crest of the transverse ridges. Due to the large dunes and the presence of the backbarrier dunes, these areas experienced less overwash penetration and most of the sediment from the beachface and dunes was deposited within the upper-shoreface. It is argued that this sediment is returned to the beachface through nearshore bar migration following the storm and that the areas with larger foredunes and backbarrier dunes have smaller rates of historical shoreline erosion compared to areas with smaller dunes and greater transfer of sediment to the washover terrace. Since the recovery of the dunes will vary depending on the availability of sediment from the washover and beachface, it is further argued that the alongshore pattern of dune morphology and the response of the island to the next extreme storm is forced by the transverse ridges and island width through alongshore variations in storm surge and overwash gradients respectively. These findings may be particularly important for coastal managers involved in the repair and rebuilding of coastal infrastructure that was damaged or destroyed during Hurricane Ivan.

Aim To provide high-resolution local, regional, national and global estimates of annual mangrove. more Aim
To provide high-resolution local, regional, national and global estimates of annual mangrove forest area from 2000 through to 2012 with the goal of driving mangrove research questions pertaining to biodiversity, carbon stocks, climate change, functionality, food security, livelihoods, fisheries support and conservation that have been impeded until now by a lack of suitable data.

Location
Global, covering 99% of all mangrove forests.

Methods
We synthesized the Global Forest Change database, the Terrestrial Ecosystems of the World database and the Mangrove Forests of the World database to extract mangrove forest cover at high spatial and temporal resolutions. We then used the new database to monitor mangrove cover at the global, national and protected area scales.

Results
Countries showing relatively high amounts of mangrove loss include Myanmar, Malaysia, Cambodia, Indonesia and Guatemala. Indonesia remains by far the largest mangrove-holding nation, containing between 26% and 29% of the global mangrove inventory with a deforestation rate of between 0.26% and 0.66% per year. We have made our new database, CGMFC-21, freely available.

Main conclusions
Global mangrove deforestation continues but at a much reduced rate of between 0.16% and 0.39% per year. Southeast Asia is a region of concern with mangrove deforestation rates between 3.58% and 8.08%, this in a region containing half of the entire global mangrove forest inventory. The global mangrove deforestation pattern from 2000 to 2012 is one of decreasing rates of deforestation, with many nations essentially stable, with the exception of the largest mangrove-holding region of Southeast Asia. We provide a standardized spatial dataset that monitors mangrove deforestation globally at high spatio-temporal resolutions. These data can be used to drive the mangrove research agenda, particularly as it pertains to monitoring of mangrove carbon stocks and the establishment of baseline local mangrove forest inventories required for payment for ecosystem service initiatives.

750, 1450 and 4550 m) that are statistically significant at the 95% confidence level. Cospectral analysis suggests that the variation in dune morphology is correlated with transverse ridges on the inner-shelf, the backbarrier cuspate headlands, and the historical and storm-related trends in shoreline change. Sections of the coast with little to no dune development before Hurricane Ivan were observed in the narrowest portions of the island (between headlands), west of the transverse ridges. Overwash penetration tended to be larger in these areas and island breaching was common, leaving the surface close to the watertable and covered by a lag of shell and gravel. In contrast, large foredunes and the backbarrier dunes were observed at the widest sections of the island (the cuspate headlands) and at crest of the transverse ridges. Due to the large dunes and the presence of the backbarrier dunes, these areas experienced less overwash penetration and most of the sediment from the beachface and dunes was deposited within the upper-shoreface. It is argued that this sediment is returned to the beachface through nearshore bar migration following the storm and that the areas with larger foredunes and backbarrier dunes have smaller rates of historical shoreline erosion compared to areas with smaller dunes and greater transfer of sediment to the washover terrace. Since the recovery of the dunes will vary depending on the availability of sediment from the washover and beachface, it is further argued that the alongshore pattern of dune morphology and the response of the island to the next extreme storm is forced by the transverse ridges and island width through alongshore variations in storm surge and overwash gradients respectively. These findings may be particularly important for coastal managers involved in the repair and rebuilding of coastal infrastructure that was damaged or destroyed during Hurricane Ivan.

750, 1450 and 4550 m) that are statistically significant at the 95% confidence level. Cospectral analysis suggests that the variation in dune morphology is correlated with transverse ridges on the inner-shelf, the backbarrier cuspate headlands, and the historical and storm-related trends in shoreline change. Sections of the coast with little to no dune development before Hurricane Ivan were observed in the narrowest portions of the island (between headlands), west of the transverse ridges. Overwash penetration tended to be larger in these areas and island breaching was common, leaving the surface close to the watertable and covered by a lag of shell and gravel. In contrast, large foredunes and the backbarrier dunes were observed at the widest sections of the island (the cuspate headlands) and at crest of the transverse ridges. Due to the large dunes and the presence of the backbarrier dunes, these areas experienced less overwash penetration and most of the sediment from the beachface and dunes was deposited within the upper-shoreface. It is argued that this sediment is returned to the beachface through nearshore bar migration following the storm and that the areas with larger foredunes and backbarrier dunes have smaller rates of historical shoreline erosion compared to areas with smaller dunes and greater transfer of sediment to the washover terrace. Since the recovery of the dunes will vary depending on the availability of sediment from the washover and beachface, it is further argued that the alongshore pattern of dune morphology and the response of the island to the next extreme storm is forced by the transverse ridges and island width through alongshore variations in storm surge and overwash gradients respectively. These findings may be particularly important for coastal managers involved in the repair and rebuilding of coastal infrastructure that was damaged or destroyed during Hurricane Ivan.

&ldquoRight&rdquo way to handle C6386 in Visual C++ (buffer overrun)?

I've got a code snippet that triggers a buffer overrun warning:

The line that triggers the warning is a[row][col] = 0 where it believes col can go negative. The warning goes away if I do any of the following:

• include && col >= 0 in the for statement (which adds an unnecessary check to a loop, since in the real code I've already checked variable bounds and know I'll never overflow)
• delete the dummy for loop in the middle of the code snippet (weird, I know, and in the real code that "dummy" loop is doing actual work)
• Change dummy to a size_t (weird, not related to the loop with the warning)
• Change n to an unsigned int (again, weird, not related to the loop in question)
• remove the odd variable from the col initializer (which would make the code incorrect)
• use a #pragma to ignore the issue (which I generally prefer to avoid)
• Check anywhere above the for loop for ranges that could lead to an overflow (I tried range checks for m , n , x , y , and even odd ).
• Change col to an unsigned type like size_t, or do any type of unsigned casting

None of the workarounds leave me feeling warm and fuzzy. I've seen warnings like these quite a bit lately, often with quirky dependencies on unrelated bits of code like the dummy for loop seen here, and almost never exposing a real issue. Is there a best practice to avoiding this type of warning?

Output buffer size of a router depends upon RTT. How?

According to a book on computer networking "a router needs an amount of buffering equal to the average round-trip time of a flow that passes through the router, multiplied by the capacity of the router’s network interface, B = RTT * C". This is the well-known rule.

My question How can this RTT be provided to a router before it is installed or how this RTT calculated beforehand? A router cannot estimate this RTT on its own as far as I know.

There is a rather persistent misconception that lock-free algorithms are faster than locking algorithms. However, that may not be true. Modern mutex implementations are extemely fast in the uncontended case, and when there is a lot of contention they use a system call that lets the kernel wait for the mutex to become unlocked. A system call definitely has a lot of overhead, but your solution is to spin 255 times. Atomic operations are not free, so in the contended case with many threads trying to get access, this might waste a lot of CPU time.

You really should try to prove your theory that a lock-free implementation is faster than one using mutexes by running benchmarks.

The name Kurna signifies "a promontory" or "a point of a mountain". [ 1 ]

References to Qurna, Gurna, Kournou, Gourna, El-Ckoor’neh, Gourne, el Abouab, El-Goor’neh or many other variants in pre-1940s literature refers to a spread out urban sprawl of housings stretching from approximately the Ramesseum (Mortuary Temple of Ramesses II) to the Mortuary Temple of Seti I on the east side of the Theban Hills, including the current place names of Sheikh ‘Adb el-Qurna , el-Assasif , el-Khokha , Dra ’Abu el-Nage ’ and Qurna .

During the 18th, 19th and 20th centuries, visitors and travelers to the area are rarely consistent in their use of the name and anything between Medinet Habu and the tombs of el-Tarif can at times be found referred to as part of a Qurna community.

A reference to the "Temple of Gourna " or similar, is in most cases a reference to the Ramesseum, to a lesser degree the Temple of Seti I and rarely it is a reference to the all but destroyed Mortuary temples of Ramesses IV, Thutmose III or Thutmose IV.