LAS files

The LAS format is widely used in the Oil and Gas industry and is short from Log ASCII Standard, in this notebook we show a short workflow to load and qc multiple files. I´m using some files donated by Geolink to the geoscience community.

The libraries

I will use to libraries:

• the tidyverse to perform data wrangling and plotting,

• petroreadr from Ravenroadresources to load the LAS files.

• skimr, to get an exccelent summary of the data.

This script can be found on my github repo

library(tidyverse)
library(petroreadr)
library(skimr)

Loading the files

The library indicates in the documentation a few options, I normally choose:

• load directly to a dataframe, which is my go to mode.\
• set the verbose option to TRUE, this is particularly useful when working with large amount of wells/logs, to have the peace of mind that the machine is working and has not hanged.

pathname <- "../../data/GEOLINK_Lithology and wells NORTH SEA/"

lasfiles <- list.files(pathname)
lasfiles <- lasfiles[grepl(".las", lasfiles)]
 
df<-read_las(file.path(pathname, lasfiles), verbose = TRUE)$data

##    + 34_10-12.las imported as <las> object
##    + 34_3-2 S.las imported as <las> object
##    + 35_3-2.las imported as <las> object
##    + 35_9-2.las imported as <las> object
##    + 35_9-9.las imported as <las> object

the loaded data

We have obtained a dataframe with all logs, and a column with the well name that makes our life easier for plotting, filtering, etc.

skim(df)

Table: Data summary

Variable type: character

Variable type: numeric

Quick view of a log

to check whether a well log is present in all wells, and compare some of their stats, the box plot is a powerful tool.

box_1 <-
  df %>%
  ggplot(aes(WELL, GR)) +
  geom_boxplot() +
  theme_bw()+
  theme(axis.text.x=element_text(angle=90,hjust=1)
        )


box_1

## Warning: Removed 16 rows containing non-finite values (stat_boxplot).

A quick view shows some negative values in well 34/10-12 and some potential outliers in well 35/9-2, we can repeat the plot cropping on a range to have a better view.

We can repeat the same for all logs at once to get a quick idea , however there are at least two points to consider when doing this:

• the size of the plot will need to be increased to properly show all logs,

• if you want to visualize a zoom of a particular log, you will need to apply a filter to the data before appliying the pivot.

• similarly, when dealing with logs that behave logarithmicaly, I dont´t know how to set a logarithmic scale to a particular log using faceting, therefore I apply a log transform before the pivoting step.

df %>% 
  mutate(log10_RDEP= log10(RDEP)) %>% 
  select(-RDEP) %>% 
  filter( GR >   0 ,
         GR  < 200 )  %>% 
  pivot_longer(!c("WELL"), names_to = "logs", values_to = "value") %>% 
  drop_na(value) %>%
  ggplot()+
  geom_boxplot(aes(WELL, value))+
  facet_wrap(~logs, scales = "free", ncol = 1) +
  theme_bw()+
  theme(axis.text.x=element_text(angle=90,hjust=1))

Density neutron plot

Just a quick plot set creating a color scale between 0 and 100, the values below 0 are filtered from the plot, but values above 100 are colored as 100.

df %>% 
  filter(GR>0) %>% 
  ggplot(aes(RHOB,NPHI, color = GR))+
  geom_point( size = 1)+
  facet_wrap(~WELL, ncol = 2)+
  xlim(1,max(df$RHOB))+
  theme_bw()+
  scale_color_gradient(low = "yellow", high = "brown",limits = c(0,100), 
                       oob = scales::squish
                       )