Skip to content

Commit

Permalink
Merge pull request #16 from nmfs-opensci/dp-patch-tut3.qmd
Browse files Browse the repository at this point in the history 
geo extend & rc_sst
  • Loading branch information
@eeholmes
eeholmes authored May 8, 2024
2 parents 8f26a01 + 47cee0f commit fc82e7f
Show file tree
Hide file tree
Showing 2 changed files with 48 additions and 73 deletions.
14 changes: 4 additions & 10 deletions tutorials/r/1-earthdatalogin.qmd
View file
Original file line number Diff line number Diff line change
@@ -1,13 +1,6 @@
---
title: Earthdata Search and Discovery
author: NOAA CoastWatch, NOAA Openscapes
date: "`r paste0(Last run format(Sys.Date(), format='%B %d %Y')) `"
output:
md_document:
variant: gfm
editor_options:
markdown:
wrap: sentence
author: Eli Holmes adapted from work by Luis Lopez and Carl Boettiger
---

::: {.callout-note title="Learning Objectives"}
Expand Down Expand Up @@ -183,10 +176,11 @@ If you get the following error:

> Error: [rast] file does not exist: /vsicurl/https://archive.podaac.earthdata.nasa.gov/podaac-ops-cumulus-protected/AVHRR_OI-NCEI-L4-GLOB-v2.1/20200115120000-NCEI-L4_GHRSST-SSTblend-AVHRR_OI-GLOB-v02.0-fv02.1.nc
It is likely because you do not have the End User Licence Agreement (EULA)/permissions to use that data set or are not properly logged in using `earthdatalogin::edl_netrc()`.
It is likely because you do not have the End User Licence Agreement (EULA)/permissions to use that data set or are not properly logged in using `earthdatalogin::edl_netrc()`. Another reason may be that your
GDAL installation is not properly handling netCDF files.
:::

Also try this example script from the `?earthdatalogin::edl_netrc` documentation:
Also try this example script from the `?earthdatalogin::edl_netrc` documentation that uses a tif file instead of netCDF.

```{r}
url <- earthdatalogin::lpdacc_example_url()
Expand Down
107 changes: 44 additions & 63 deletions tutorials/r/3-extract-satellite-data-within-boundary.qmd
View file
Original file line number Diff line number Diff line change
@@ -1,67 +1,46 @@
---
title: Extract data within a boundary
author: NOAA CoastWatch, NOAA Openscapes
date: "`r paste0(Last run format(Sys.Date(), format='%B %d %Y')) `"
output:
md_document:
variant: gfm
editor_options:
markdown:
wrap: sentence
author: 'NOAA CoastWatch, NOAA Openscapes'
---

::: {.callout-note title="Learning Objectives"}

1. How to access and download sea surface temperature from NASA Earthdata
2. How to apply shapefiles as masks to satellite data
3. How to compute monthly average sea surface temperature
1. How to access and download sea surface temperature from NASA Earthdata
2. How to apply shapefiles as masks to satellite data
3. How to compute monthly average sea surface temperature
:::


## Summary

In this example, we will utilize the `earthdatalogin` R package to retrieve sea surface temperature data from NASA Earthdata.

The `earthdatalogin` package simplifies the process of discovering and accessing NASA Earth science data.


This example is adapted from the NOAA CoastWatch Satellite Data Tutorials. To explore the full range of
tutorials on accessing and utilizing oceanographic satellite data,
visit the [NOAA CoastWatch Tutorial Github repository.](https://github.com/coastwatch-training/CoastWatch-Tutorials)
In this example, we will utilize the earthdatalogin R package to retrieve sea surface temperature data from [NASA Earthdata search](https://search.earthdata.nasa.gov/search). The `earthdatalogin` package simplifies the process of discovering and accessing NASA Earth science data.

This example is adapted from the NOAA CoastWatch Satellite Data Tutorials. To explore the full range of tutorials on accessing and utilizing oceanographic satellite data, visit the [NOAA CoastWatch Tutorial Github repository.](https://github.com/coastwatch-training/CoastWatch-Tutorials)

For more on `earthdatalogin` visit the
[`earthdatalogin` GitHub](https://github.com/boettiger-lab/earthdatalogin/)
page and/or the [`earthdatalogin` documentation](https://boettiger-lab.github.io/earthdatalogin/) site.
Be aware that `earthdatalogin` is under active development.
For more on `earthdatalogin` visit the [`earthdatalogin` GitHub](https://github.com/boettiger-lab/earthdatalogin/) page and/or the [`earthdatalogin` documentation](https://boettiger-lab.github.io/earthdatalogin/) site. Be aware that `earthdatalogin` is under active development and that we are using the development version on GitHub.

## Prerequisites

An Earthdata Login account is required to access data from NASA Earthdata.
Please visit <https://urs.earthdata.nasa.gov> to register and manage
your Earthdata Login account. This account is free to create and
only takes a moment to set up.
The tutorials today can be run with the guest Earthdata Login that is in `earthdatalogin`.
However, if you will be using the NASA Earthdata portal more regularly, please register for an
Earthdata Login account. Please <https://urs.earthdata.nasa.gov> to register and manage your
Earthdata Login account. This account is free to create and only takes a moment to set up.

*Note: See the [Earthdata login set-up tab](https://nmfs-opensci.github.io/EDMW-EarthData-Workshop-2024/content/02-earthdata.html) (in left nav bar) for instructions on getting set up on your own computer.*
### Import Required Packages

*Note: See the set-up tab (in left nav bar) for instructions on getting set up on your own computer, but
be aware that getting it is common to run into trouble getting GDAL set up properly to handle
netCDF files. Using a Docker image (and Python) is often less aggravating.*

## Datasets used
__GHRSST Level 4 AVHRR_OI Global Blended Sea Surface Temperature Analysis (GDS2) from NCEI__
This NOAA blended SST is a moderate resolution satellite-based gap-free
sea surface temperature (SST) product. We will use the daily data.
https://cmr.earthdata.nasa.gov/search/concepts/C2036881712-POCLOUD.html

__Longhurst Marine Provinces__
The dataset represents the division of the world oceans
into provinces as defined by Longhurst (1995; 1998; 2006).
This division has been based on the prevailing role of physical
forcing as a regulator of phytoplankton distribution.
**GHRSST Level 4 AVHRR_OI Global Blended Sea Surface Temperature Analysis (GDS2) from NCEI**\
This NOAA blended SST is a moderate resolution satellite-based gap-free sea surface temperature (SST) product. We will use the daily data. https://cmr.earthdata.nasa.gov/search/concepts/C2036881712-POCLOUD.html

**Longhurst Marine Provinces**\
The dataset represents the division of the world oceans into provinces as defined by Longhurst (1995; 1998; 2006). This division has been based on the prevailing role of physical forcing as a regulator of phytoplankton distribution.

The Longhurst Marine Provinces dataset is available online
(https://www.marineregions.org/downloads.php) and
within the shapes folder associated with this repository.
For this exercise we will use the Gulf Stream province (ProvCode: GFST)
The Longhurst Marine Provinces dataset is available online (https://www.marineregions.org/downloads.php) and within the shapes folder associated with this repository. For this exercise we will use the Gulf Stream province (ProvCode: GFST)

![../images/longhurst.png](../images/longhurst.png)
![](../images/longhurst.png)

## Load packages

Expand All @@ -72,10 +51,11 @@ library(sf)
library(ggplot2)
```

## Load boundary coordinates
## Load boundary coordinates

The shapefile for the Longhurst marine provinces includes a list of regions.\
For this exercise, we will only use the boundary of one province, the Gulf Stream region ("GFST").

The shapefile for the Longhurst marine provinces includes a list of regions.
For this exercise, we will only use the boundary of one province, the Gulf Stream region ("GFST").

```{r read province boundaries from shapefiles}
# Set directory path for shapefile
Expand All @@ -94,7 +74,7 @@ xcoord <- st_coordinates(GFST)[,1]
ycoord <- st_coordinates(GFST)[,2]
```

## Search data from NASA Earthdata with the dataset unique name and coordinates/dates
## Search data from NASA Earthdata with the dataset unique name and coordinates/dates

First, connect to NASA Earthdata with no credentials

Expand Down Expand Up @@ -125,6 +105,9 @@ results <- edl_search(
# Check number of files
length(results)
```

There are `r length(results)` files.

## Apply shapefiles as mask to satellite data

```{r}
Expand All @@ -149,8 +132,8 @@ plot(GFST,col='red')
# Mask SST with the GFST boundaries
masked_rc <- mask(ras_sst, GFST)
# Visualize the SST in GFST Province
plot(masked_rc)
# Visualize the SST in GFST Province and crop to the GFST extent
plot(masked_rc, ext = GFST)
```

::: {.callout-note title="Troubleshooting"}
Expand All @@ -164,17 +147,18 @@ If you get the following error:
## Compute monthly average of SST

We will construct a data cube to compute monthly average for sea surface temperature data within the boundary.
To minimize data loading times, the first 10 results, which correspond to approximately two months
of data, will be used for this exercise.

To minimize data loading times, the first 10 results, which correspond to approximately two months of data, will be used for this exercise.


Select the SST results for end of Jan and beginning of Feb
```{r}
# Select the first 10 SST results
ras_all <- terra::rast(results[c(1:10)], vsi = TRUE)
ras_all <- terra::rast(results[c(20:40)], vsi = TRUE)
```

# Trim the SST data to the boundaries of GFST
Trim the SST data to the boundaries of GFST
```{r}
rc_all <- terra::mask(ras_all, GFST)
```
# SST data
rc_sst <- rc_all["analysed_sst", ]

Expand All @@ -184,16 +168,16 @@ year_month <- function(x) {
}

# Convert time to Year-month format for aggregation
ym <- year_month(rc_all)
ym <- year_month(rc_sst)

# Compute raster mean grouped by Year-month
monthly_mean_rast <- terra::tapp(rc_all, ym, fun = mean)
monthly_mean_rast <- terra::tapp(rc_sst, ym, fun = mean)

# Compute mean across raster grouped by Year-month
monthly_means <- global(monthly_mean_rast, fun = mean, na.rm=TRUE)
```
## Convert raster into data frame
## Convert raster into data frame
```{r}
# Convert raster into data.frame
Expand All @@ -206,12 +190,9 @@ monthly_means_df$year_month <- sub("X", "", rownames(monthly_means_df))
## Plot monthly mean of sea surface temperature within GFST province

```{r}
# Plot monthly mean
ggplot(data = monthly_means_df, aes(x = year_month, y = mean, group = 1)) +
geom_line() +
geom_point() +
xlab("Year.Month") +
ylab("Mean SST (F)")
```

0 comments on commit fc82e7f

Please sign in to comment.