tidync-filtrate.R

cls <- c("#440154FF", "#471164FF", "#482071FF", "#472E7CFF", "#443C84FF", 
"#3F4889FF", "#39558CFF", "#34618DFF", "#2F6C8EFF", "#2A768EFF", 
"#26818EFF", "#228B8DFF", "#1F958BFF", "#1FA088FF", "#24AA83FF", 
"#30B47CFF", "#41BD72FF", "#57C666FF", "#6FCF57FF", "#8AD647FF", 
"#A7DB35FF", "#C4E021FF", "#E2E418FF", "#FDE725FF")
library(ncdump)
f <- "/rdsi/PRIVATE/raad/data/eclipse.ncdc.noaa.gov/pub/OI-daily-v2/NetCDF/2017/AVHRR/avhrr-only-v2.20170502_preliminary.nc"
x <- NetCDF(f)
x

## [1] "Variable: sst"
## [1] "(anom, err, ice)"
## [1] "Dimension:"
##   name length unlimited
## 1 time      1     FALSE
## 2 zlev      1     FALSE
## 3  lat    720     FALSE
## 4  lon   1440     FALSE

nctive(x)

## [1] "sst"

## push sst to the front for an extraction
x <- activate(x, "sst")
hyper_slab <- filtrate(x, lon = lon > 100, lat = lat < 30)

## Joining, by = ".variable_"

## Joining, by = ".dimension_"

## Joining, by = "id"

## it's alive, test the extraction
library(ncdf4)
nc <- nc_open(f)
library(dplyr)

## 
## Attaching package: 'dplyr'

## The following objects are masked from 'package:stats':
## 
##     filter, lag

## The following objects are masked from 'package:base':
## 
##     intersect, setdiff, setequal, union

sst <- ncvar_get(nc, "sst", start = bind_rows(hyper_slab)$start, count = bind_rows(hyper_slab)$count)
image(sst, col = cls)

## push a different var to the front
x <- activate(x, "anom")
hyper_slab <- filtrate(x, lon = between(lon, 147, 250), lat = between(lat, -42, 20))

## Joining, by = ".variable_"

## Joining, by = ".dimension_"

## Joining, by = "id"

anom <- ncvar_get(nc, "sst", start = bind_rows(hyper_slab)$start, count = bind_rows(hyper_slab)$count)
image(anom, col = cls)