In camposfa/paceR: Tools for working with the PACE Database

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>",
  fig.path = "PHENO-"
)

Obtaining and analyzing phenology data

library(paceR)
load_pace_packages()

I assume that you have already set up the connections to the database.

  # Connect to monkey database
  pace_db <- src_mysql(group = "PACE", user = "camposf", dbname = "monkey", password = NULL)

  # Connect to paceR database  
  paceR_db <- src_mysql(group = "PACE", user = "camposf", dbname = "paceR", password = NULL)

  options(dplyr.width = Inf)

Santa Rosa Phenlology Data

First get Santa Rosa phenology data from the database using the saved View.

ph <- getv_Phenology(paceR_db)

# Filter to obtain Santa Rosa data only
ph <- ph %>% filter(Project == "SR")

Fortunately, Santa Rosa doesn't mix categorical scores, percents, and counts. Therefore, we can drop the percents and counts (which aren't used) and focus only on the scores.

# Remove unnecessary columns
ph <- ph %>% select(-PhenologyPercent, -PhenologyCount, -ScientificName, -RecordDate)

# Let's look at the data
ph

The data are stored in "long" format in the database, where each distinct measurement is a row. If you want to see them in a "wide" format, where all measurements for a given tree/session are in one row (like the spreadsheet in which they are collected), we can reshape it like so:

# First unite the "FoodPart" and "Measurement" columns
ph_wide <- ph %>% unite(FoodPartMeasurement, c(FoodPart, Measurement))

# Now spread PhenologyScore using FoodPartMeasurement as the key
ph_wide <- ph_wide %>% spread(FoodPartMeasurement, PhenologyScore)

# Do a bit of column rearranging
ph_wide <- ph_wide %>% select(1:6, 9:15, ResearcherName, Comments)

# Look at data
ph_wide