What is the scope of NEON lakes data?

Detailed setup files and runnable R scripts for reproducing this blog post are at: https://github.com/jsta/neon_lakes

The NEON documentation lists lakes among their targeted ecosystems. However, NEON provides so much data it can be difficult to get a grasp of their lake data apart from data collected in other ecosystems. Here, I show the scope of NEON lake data and how to get scripted access to it. In particular, I attempt to answer the question: How much lake data is available through NEON and what types of data are provided?

Dependencies

Before we begin, let’s set up our R environment where the package dependencies are:

• geoNEON

• neonUtilities

Walkthrough

According to the NEON documentation, there are 7 NEON lake sites. We can start by mapping the location of these sites (we’ll exclude TOOK in Alaska). We’ll pull the coordinates for each site using the geoNEON function def.extr.geo.os:

neon_lakes  <- data.frame( # excluding `TOOK` in Alaska
    siteID = c("CRAM", "SUGG", "BARC", "PRPO", "LIRO", "PRLA"),
      stringsAsFactors = FALSE) 
neon_lakes <- geoNEON::def.extr.geo.os(neon_lakes, locCol = "siteID")

NEON provides a spreadsheet of data availability by type, site, and date where years represent the earliest year of data collection. Below, I’ve provided a rendered table of this excel sheet filtered by lake sites. The table is sorted vertically (according to variable name) from least to most data available and horizontally (according to lake site) from shortest to longest monitoring period. For example, SUGG secchi depth data is quite extensive whereas DNA barcoding data is fairly new and limited.

Let’s pull data values for an arbitrary data product by finding its product ID code in the NEON Data Product Catalog and feeding this ID to the neonUtilities function loadByProduct. Here, I pull Secchi depth measurements from each of our 6 lake sites.

secchi <- neonUtilities::loadByProduct(dpID = "DP1.20252.001",
                                         site = neon_lakes$siteID,
                                         check.size = FALSE)

It looks like although there are not huge numbers of NEON lake sites, the coverage in terms of monitoring variables within a given site is quite comprehensive. Also, the NEON folks have done a great job making their data available via scripted access.