Tutorial
NEON Mobile Deployment Platform Data: Download and Explore
Authors: Cove Sturtevant, Claire Lunch, David Durden, Chris Florian
Last Updated: Sep 18, 2025
This tutorial provides an introduction to working with NEON Mobile Deployment
Platform (MDP) data, using the neonMDP and neonUtilities R packages. If you are new to NEON data, we also recommend more general tutorials, including the Download and Explore tutorial
and the introduction to NEON eddy flux data.
A valid login to a NEON MDP data bucket is required to use this tutorial.
A key difference between public NEON data and MDP data is that MDP data are not retrieved from the NEON Data Portal. Rather, data are accessed directly from a Google Cloud Storage (GCS) bucket with a more complex directory structure than data packages downloaded from the Portal. This tutorial shows a user how to download the data from the bucket to a local data store and load it into the R computing environment for exploration. At that point the data formatting matches other NEON data retrieval methods, and users can turn to standard NEON tutorials for further guidance. Pointers to particularly relevant tutorials are also provided.
This tutorial ends with instructions to write the data loaded into the R environment to .csv format should the user desire to work with the data in another application in a more traditional format.
Learning Objectives
After completing this tutorial you will be able to:
- Download, compile, and export MDP data
Things You’ll Need To Complete This Tutorial
Bucket login
If you are working on a NEON MDP project and do not already have a login to the GCS data bucket, consult the MDP project PI and the NEON Research Support Services contact.
R Programming Language
You will need a current version of R to complete this tutorial. We also recommend the RStudio IDE to work with R.
R Packages to Install
Prior to starting the tutorial ensure that the following packages are installed.
-
devtools:
install.packages("devtools") -
neonMDP:
devtools::install_github('NEONScience/NEON-mobile-deployment/neonMDP') -
neonUtilities:
install.packages("neonUtilities") -
googleCloudStorageR:
install.packages("googleCloudStorageR")
Load the packages.
library(neonUtilities)
library(neonMDP)
library(googleCloudStorageR)
Working Directory
This lesson will download MDP data to your working directory. Set the working directory, modifying the file path in the code below as needed for your computer.
setwd("~/data/NEON/MDP")
Download Instrument Data from Google Cloud Storage Bucket
First, we will download all Instrument System (IS) data from the GCS bucket. IS data includes all automated sensors except for the Surface-Atmosphere Exchange data, which will be downloaded in a separate section below.
Change the bucket in the code below to the bucket set up for your MDP data, then run this section once to download all existing IS data from the bucket. Note that the code will error if data have already been downloaded. To retrieve new data each month, see the "get new data" section below.
setGcsEnv(bucket='neon-rss-md04')
downloadMDP('Publication')
Stack IS Data
NEON data are provided in product-site-month increments.
The following code will stack (compile) the full time-series of IS products.
The stackfromStore() function stacks data downloaded to the local environment,
which we did earlier.
Examples for three products are provided below. You can see which products are
available by looking in the "Publication" folder of the working directory.
If there are data from more than one site in the working directory, change
the site= parameter in the code snippets below from "all" to the 4-character
MDP site code (e.g. "MD04") that you want to explore.
If data in the MDP bucket have been republished, stackFromStore() defaults to
the most recently published data for each time period. Generally this is the
right choice, to get the most up-to-date data, but use the pubdate= input to
stack older versions.
See the Help page for the stackFromStore() function for descriptions of other
options you may want to change (e.g. the date range or averaging interval to
stack).
# Single-aspirated Air Temperature
saat <- stackFromStore(dpID='DP1.00002.001',
filepaths=getwd(),
site='all',
startdate=NA, enddate=NA,
package='expanded')
# Relative Humidity
rh <- stackFromStore(dpID='DP1.00098.001',
filepaths=getwd(),
site='all',
startdate=NA, enddate=NA,
package='basic')
# Water Quality
wq <- stackFromStore(dpID='DP1.20288.001',
filepaths=getwd(),
site='all',
startdate=NA, enddate=NA,
package='expanded')
You will see messages indicating the Release tag could not be found for the stacked tables. This is expected for MDP data, since they do not follow the standard NEON data Release paradigm. The messages can be ignored.
The outputs are named lists containing data frames of time-series data along with other tables containing metadata.
Note that some of the metadata may be irrelevant to your MDP project - for
example, the issueLog table contains data product issues related to sites in
the NEON project design.
# 30-minute single-aspirated air temperature (1st 10 rows & columns)
saat$SAAT_30min[1:10,1:10]
## domainID siteID horizontalPosition verticalPosition startDateTime endDateTime tempSingleMean
## 1 D14 MD04 000 010 2025-03-27 00:00:00 2025-03-27 00:30:00 NA
## 2 D14 MD04 000 010 2025-03-27 00:30:00 2025-03-27 01:00:00 NA
## 3 D14 MD04 000 010 2025-03-27 01:00:00 2025-03-27 01:30:00 NA
## 4 D14 MD04 000 010 2025-03-27 01:30:00 2025-03-27 02:00:00 NA
## 5 D14 MD04 000 010 2025-03-27 02:00:00 2025-03-27 02:30:00 NA
## 6 D14 MD04 000 010 2025-03-27 02:30:00 2025-03-27 03:00:00 NA
## 7 D14 MD04 000 010 2025-03-27 03:00:00 2025-03-27 03:30:00 NA
## 8 D14 MD04 000 010 2025-03-27 03:30:00 2025-03-27 04:00:00 NA
## 9 D14 MD04 000 010 2025-03-27 04:00:00 2025-03-27 04:30:00 NA
## 10 D14 MD04 000 010 2025-03-27 04:30:00 2025-03-27 05:00:00 NA
## tempSingleMinimum tempSingleMaximum tempSingleVariance
## 1 NA NA NA
## 2 NA NA NA
## 3 NA NA NA
## 4 NA NA NA
## 5 NA NA NA
## 6 NA NA NA
## 7 NA NA NA
## 8 NA NA NA
## 9 NA NA NA
## 10 NA NA NA
# variable descriptions for the single-aspirated air temperature data
saat$variables_00002[1:10,]
## table fieldName description dataType units
## <char> <char> <char> <char> <char>
## 1: SAAT_1min domainID Unique identifier of the NEON domain string <NA>
## 2: SAAT_1min siteID NEON site code string <NA>
## 3: SAAT_1min horizontalPosition Index of horizontal location at a NEON site string <NA>
## 4: SAAT_1min verticalPosition Index of vertical location at a NEON site string <NA>
## 5: SAAT_1min startDateTime Date and time at which a sampling is initiated dateTime <NA>
## 6: SAAT_1min endDateTime Date and time at which a sampling is completed dateTime <NA>
## 7: SAAT_1min tempSingleMean Arithmetic mean of single aspirated air temperature real celsius
## 8: SAAT_1min tempSingleMinimum Minimum single aspirated air temperature real celsius
## 9: SAAT_1min tempSingleMaximum Maximum single aspirated air temperature real celsius
## 10: SAAT_1min tempSingleVariance Variance in single aspirated air temperature real celsiusSquared
## downloadPkg pubFormat primaryKey categoricalCodeName
## <char> <char> <char> <char>
## 1: appended by stackByTable <NA> N
## 2: appended by stackByTable <NA> N
## 3: appended by stackByTable <NA> N
## 4: appended by stackByTable <NA> N
## 5: basic yyyy-MM-dd'T'HH:mm:ss'Z'(floor) <NA> <NA>
## 6: basic yyyy-MM-dd'T'HH:mm:ss'Z'(floor) <NA> <NA>
## 7: basic *.####(round) <NA> <NA>
## 8: basic *.####(round) <NA> <NA>
## 9: basic *.####(round) <NA> <NA>
## 10: basic *.####(round) <NA> <NA>
For more information about naming conventions, data tables, and metadata, see the Download and Explore tutorial.
Download Surface-Atmosphere Exchange Data from Google Cloud Storage Bucket
Now we will download all Surface-Atmosphere Exchange (SAE) data from the GCS bucket.
Change the bucket in the code below to the bucket set up for your MDP data, then run this section once to download all existing SAE data from the bucket. Note that the code will error if data have already been downloaded. To retrieve new data each month, see the "get new data" section below.
setGcsEnv(bucket='neon-rss-md04')
downloadMDP('ods/dataproducts/DP4')
Stack SAE Data
The following code will stack (compile) SAE products providing a complete
time-series for the downloaded MDP site. The stackFromStore() function utilizes
the stackEddy() function which is described in great detail in the
Introduction to working with NEON eddy flux data tutorial.
The MDP data files are in the same format as described in the tutorial. Below we
show examples of stacking the Level 4 flux data and Level 1 isotopic ratios.
If there are data from more than one site in the working directory, change
the site= parameter in the code snippets below from "all" to the 4-character
MDP site code (e.g. "MD04") that you want to explore. See the Help pages for the
stackFromStore() function and the stackEddy() function for descriptions of
other options you may want to change.
# NSAE data
flux <- stackFromStore(dpID='DP4.00200.001',
filepaths=paste(getwd(), 'ods', sep='/'),
site='all', startdate=NA, enddate=NA,
package='basic', level='dp04')
# CO2 isotope data
iso <- stackFromStore(dpID='DP4.00200.001',
filepaths=paste(getwd(), 'ods', sep='/'),
site='all', startdate=NA, enddate=NA,
package='basic', level='dp01',
timeIndex=30, var='dlta13CCo2')
You may see a warning saying something like
no non-missing arguments to max; returning -Inf. This is related to the daily
basic package files in the MDP bucket. These files are ignored in stacking
using stackFromStore(); consult NEON if you need to use the daily basic files.
Otherwise you can ignore this warning.
The resultant data lists will include the variables and objDesc tables to
help you interpret the data by describing term definitions and units for
variables.
# flux data variables (1st 10 rows)
flux$variables[1:10,]
## category system variable stat units
## 1 data fluxCo2 nsae timeBgn NA
## 2 data fluxCo2 nsae timeEnd NA
## 3 data fluxCo2 nsae flux umolCo2 m-2 s-1
## 4 data fluxCo2 stor timeBgn NA
## 5 data fluxCo2 stor timeEnd NA
## 6 data fluxCo2 stor flux umolCo2 m-2 s-1
## 7 data fluxCo2 turb timeBgn NA
## 8 data fluxCo2 turb timeEnd NA
## 9 data fluxCo2 turb flux umolCo2 m-2 s-1
## 10 data fluxH2o nsae timeBgn NA
# CO2 isotope data (1st 10 rows & columns)
iso$MD04[1:10,1:10]
## horizontalPosition verticalPosition timeBgn timeEnd data.isoCo2.dlta13CCo2.mean
## 1 000 010 2025-04-01 00:00:00 2025-04-01 00:29:59 NaN
## 2 000 010 2025-04-01 00:30:00 2025-04-01 00:59:59 NaN
## 3 000 010 2025-04-01 01:00:00 2025-04-01 01:29:59 NaN
## 4 000 010 2025-04-01 01:30:00 2025-04-01 01:59:59 NaN
## 5 000 010 2025-04-01 02:00:00 2025-04-01 02:29:59 NaN
## 6 000 010 2025-04-01 02:30:00 2025-04-01 02:59:59 NaN
## 7 000 010 2025-04-01 03:00:00 2025-04-01 03:29:59 NaN
## 8 000 010 2025-04-01 03:30:00 2025-04-01 03:59:59 NaN
## 9 000 010 2025-04-01 04:00:00 2025-04-01 04:29:59 NaN
## 10 000 010 2025-04-01 04:30:00 2025-04-01 04:59:59 NaN
## data.isoCo2.dlta13CCo2.min data.isoCo2.dlta13CCo2.max data.isoCo2.dlta13CCo2.vari data.isoCo2.dlta13CCo2.numSamp
## 1 NaN NaN NA 0
## 2 NaN NaN NA 0
## 3 NaN NaN NA 0
## 4 NaN NaN NA 0
## 5 NaN NaN NA 0
## 6 NaN NaN NA 0
## 7 NaN NaN NA 0
## 8 NaN NaN NA 0
## 9 NaN NaN NA 0
## 10 NaN NaN NA 0
## qfqm.isoCo2.dlta13CCo2.qfFinl
## 1 1
## 2 1
## 3 1
## 4 1
## 5 1
## 6 1
## 7 1
## 8 1
## 9 1
## 10 1
Refer to the Introduction to working with NEON eddy flux data tutorial for further details.
Get New Data
To retrieve new data published to the bucket each month, change the bucket in the code below to the bucket set up for your MDP data, then run this section.
setGcsEnv(bucket='neon-rss-md04')
updateMDP('Publication')
updateMDP('ods/dataproducts/DP4')
Explore Data by Accessing other NEON Tutorials
After stacking the IS data, you can transfer to the
Explore Temperature data section in the Get Started with NEON Data series of tutorials. The tutorial guides you through the data and metadata contents of the list output from the stackFromStore() function for IS products, including how to interpret quality flags as well as do basic plotting and manipulation of data tables.
There are several SAE related tutorials. The aforementioned Introduction to working with NEON eddy flux data is a great starting point. Another tutorial that may be of interest is the Exploring diel carbon flux cycles, which provides a more comprehensive exploration of the carbon fluxes by looking at diel patterns and highlighting how to utilize NEON quality flags to evaluate data quality.
Save Data in .csv Format
If preferred, IS time-series data compiled with the stackFromStore() function can easily be written to .csv format. The following example code will write the 30-minute single-aspirated air temperature data to your working directory.
write.csv(saat$SAAT_30min,
file="SAAT_30min_stacked.csv",
row.names=F)
Writing out the stacked SAE data to .csv is similarly easy. The following example code will write the 30-minute CO2 isotope data to the working directory. Change the "MD04" site code in iso$MD04 in the code below to the 4-character site code for your MDP.
write.csv(iso$MD04,
file="co2_isotopes_30min_stacked.csv",
row.names=F)