Statistics for Geosciences

Prepared by: Ghozian Islam Karami

This R markdown is just an example how we can utilize R for statistics analysis by using porosity and permeability data. We will Install the library, and load to the R environment. We will also load the porosity-permeability data from my github page : R for Geosciences

#install the library requirement
if (!require("pacman")) install.packages("pacman")

Loading required package: pacman

pacman::p_load(tidyverse, pastecs, ggpubr)

#load the dataset
df <- read.csv("https://raw.githubusercontent.com/ghoziankarami/Rforgeosciences/main/Porosity-Permeability.csv")

Exploratory Data Analysis (EDA)

#to define the first rows of the data
head(df)

NA

Univariate Data Statistics

#How to generate descriptive statistics
summary(df)

   Depth..m.       Porosity       Permeability  
 Min.   :1500   Min.   : 2.840   Min.   : 30.0  
 1st Qu.:1640   1st Qu.: 6.420   1st Qu.:150.0  
 Median :1780   Median : 7.890   Median :205.0  
 Mean   :1772   Mean   : 8.135   Mean   :222.5  
 3rd Qu.:1910   3rd Qu.: 9.890   3rd Qu.:310.0  
 Max.   :2040   Max.   :12.550   Max.   :430.0  

# Compute descriptive statistics
library(tidyverse)
library(pastecs)
df %>% stat.desc() %>% round(2)

#Define the variable
Porosity <- df$Porosity
Permeability <- df$Permeability

#How to generate a histogram
hist(Porosity, main="Porosity Histogram")

hist(Permeability, main="Permeability Histogram")

#How to generate a boxplot
boxplot(Porosity, main="Porosity Boxplot")

boxplot(Permeability, main="Permeability Boxplot")

#How to combining plots
par(mfrow=c(2,2))

hist(Porosity, main="Porosity Histogram")
hist(Permeability, main="Permeability Histogram")

boxplot(Porosity, main="Porosity Boxplot")
boxplot(Permeability, main="Permeability Boxplot")

#To generate cdf

a<- ggecdf(df, x = "Permeability")
b<- ggecdf(df, x = "Porosity")
library(patchwork)
a + b + plot_annotation(title = "Cumulative Distribution Function") #side by side

Bivariate Data Statistics

#Scatter Plot
plot(Permeability ~ Porosity, dat = df)
regression <- lm(Permeability ~ Porosity, data=df)
abline(regression)

#compute the summary of the regression line
summary(regression)

Call:
lm(formula = Permeability ~ Porosity, data = df)

Residuals:
   Min     1Q Median     3Q    Max 
-94.93 -46.88 -12.99  36.70 155.77 

Coefficients:
            Estimate Std. Error t value Pr(>|t|)    
(Intercept)  -29.008     34.175  -0.849      0.4    
Porosity      30.918      4.043   7.646 8.62e-10 ***
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Residual standard error: 64.89 on 47 degrees of freedom
Multiple R-squared:  0.5544,    Adjusted R-squared:  0.5449 
F-statistic: 58.47 on 1 and 47 DF,  p-value: 8.624e-10

#Plot the regression
plot(regression)

#Using GGplot

regplot <- ggplot(df, aes(y=Permeability, x=Porosity)) + geom_point() + geom_smooth(method = "lm") + stat_regline_equation() + theme_bw() + labs(title="Simple regrssion of Permeability and Porosity")

regplot

`geom_smooth()` using formula 'y ~ x'