Flowering phenology under climate warming
Data files
Sep 25, 2023 version files 32.47 KB
-
phenology_raw_data_Dryad.zip
-
README.md
Abstract
Owing to global climate warming, biological processes will be disrupted, causing phenological mismatches and mistimings. In Japan's cool temperate region of Hokkaido Island, warmer and shorter winters will affect plant flowering phenology. Rapid change in the flowering phenology led to the hypothesis that shorter winters would result in earlier flowering. The flowering phenology of 611 species was recorded for nine years (2013–2021), and 9 species were selected for analysis. The accumulated degree hour, which is the total value of the hourly aerial temperature above zero, was calculated as the effective aerial temperature for plant flowering phenology. The results indicated that winters are getting shorter, and the first flowering day of two plant species occurred earlier. However, the accumulated degree hours for the plant species have not significantly changed over the nine years. If climate warming continues, it will have diverse, complex, and unpredictable effects on various individual organisms and biological relationships among species. Detailed studies are needed to link climate change predictions to the predicted degree of mismatch in species interactions and networks.
README: Flowering phenology under climate warming
Description of the data and file structure
Raw data as three R source files and two CSV files.
Climate data (1961–2021) were collected from the Japan Meteorological Agency of Asahikawa (JMA, https://www.data.jma.go.jp/obd/stats/etrn/) using the R packages ‘polite’ and ‘XML.’
Thus, data was primarily produced as R source file and saved as 'dataframe' object.
I will not publish the scraping scripts to avoid too-much-easy access to the JMA.
All of the CSV files are raw data, but produced through the 'dataframe' above, and not directly used in the R analysis.
How to run the analysis
R script for analysis is included. You need the font "Osaka" to draw graphs.
If you do not want to use "Osaka", you can change par(family="Osaka") into another font.
Please place the three source files at your proper directory for R analysis.
Files seppen.R and rec.snow.long2021.R are the source files and used in R script file "ESJ69_PB_eng.v2.eng.R".
And you should change the file path in the script file.
e.g., you should change the three lines load(file="/rec.snow.long2021.R") , seppen.R, and SList.ssp.sel1.R.
You can check them using the function 'ls()'
Details on source files
The unit of temperature is Celsius.
Day of the year is represented as 'yday'.
The object 'rec.snow.long2021' contains snowfall and snow depth (cm) day by day 1961-2021.
The values "snow" and "snow.max" denote snowfall and snow depth, respectively.
The object "kousetu" and "sekisetu2" are generated in the R script (line 54 and line 67) and denotes monthly total of snowfall and monthly maximum snowfall per year (cm), respectively.
However, the CSV files generated from these objects are not directly used in the analysis.
Names of the species and the species number are column one and five, respectively.
Species names are abbreviated as in my NJB article (see its DOI in my Dryad page.)
The object 'SList.ssp.sel1' contains data on flowering day (day of the year).
Flowering day and early.day.of.flower are pooled as one value.
All the SITE were pooled and analysed.
The column 'kaikabi' denotes 'yday' of the first flowering and is used in Fig 4a of my NJB article.
The column 'kaika.sekisan0' is important and used in Fig 4b of my NJB article.
The values of accumulated degree hour (ADH) above zero degrees in my article.
The column 'kaika.sekisan' is the value of accumulated degree hour (ADH) above five degrees and not directly used.
The column 'mean.sekisan' and ''mean.sekisan0'' is the value of accumulated degree day (ADD) above five and zero degrees, although not directly used.
ADD is calculated using calculated daily mean temperature.
The column "sekisetu.zero" is calculated yday of snowmelt (zero snow depth day) , referred from the object "seppen," and used in Fig 4c of my NJB article.
Columns of "line..." just represent line types in the graphs according to p-value of the regression (see Figure legend of NJB article).
Sharing/Access information
Data can be used under CC0