Formatting the data for crestr

Manuel Chevalier

2021-12-03

The central element of the package: the crestObj object

In crestr, all the CREST-related data are stored within a single S3 object of the class crestObj. Most functions of the package will take a crestObj as their main input and will return an updated version of the object. In practice, a crestObj is a nested list that contains five sub-lists, each one grouping a specific type of information, such as the calibration data, the fitted climate responses and the reconstructions (see Fig. 1). Wrapper functions have been implemented to manipulate and/or modify the information contained in a crestObj so that, users are never expected to manually modify their crestObj — even if it is possible. The five sub-lists contain the following information:


**Fig. 1** Structure of a crestObj, here called 'rcnstrctn', with the five main sub-lists in colour. For simplicity, lists with many elements ('tax' or 'clim') are represented with double framed boxes. The unframed terminal nodes on the right-hand side of each branch are simple R objects, such as numbers, strings, vectors or data frames. The names of the functions that modify these obsjects are indicated on the right.

Fig. 1 Structure of a crestObj, here called ‘rcnstrctn’, with the five main sub-lists in colour. For simplicity, lists with many elements (‘tax’ or ‘clim’) are represented with double framed boxes. The unframed terminal nodes on the right-hand side of each branch are simple R objects, such as numbers, strings, vectors or data frames. The names of the functions that modify these obsjects are indicated on the right.


Formatting the input data for crestr

Four different input data files are compatible with crestr. However, only one up to three of these inputs may be required for any application depending on the modelling objectives and assumptions. These files can be prepared outside the R environment and imported using standard R functions.


The fossil data (df)

The df data frame is only required if crestr is used for reconstucting climate and can be omitted if the objective is to model the climate response(s) of different taxa. df should be a data frame with the different samples entered as rows, with either the age, depth or sample ID as first column and the fossil data in the subsequent columns. df can countain raw counts, percentages, presence/absence (1s and 0s) or even relative weights to be used in the reconstruction. In the case study presented here, 215 terrestrial and aquatic pollen taxa were observed in 181 samples, so that df is a data frame with 216 columns (the age of the samples followed by the pollen taxa) and 181 rows containing pollen percentages of terrestrial taxa.


The proxy-species equivalency (PSE) table

The PSE data frame is required to use the gbif4crest calibration dataset. It is used to associate fossil taxon names to the species distributions of various species stored in the TAXA and DISTRIB_QDGC tables (Fig. 2 here). When all the fossil taxa are identified at the species level, the PSE table should be a simple data frame with one row per taxon. In most cases, however, fossil taxa are identified at a higher taxonomic level (sub-genus, genus, sub-family, family) and these varying levels of identification should be encoded in the PSE file to link one or more (groups of) species to their common fossil taxon name.

A PSE file is composed of five columns (see Table 1). The first one (Level) contains an integer that indicates the level of taxonomic resolution of the row (1 for Family, 2 for Genus, 3 for Species and 4 for taxa that should be excluded from the reconstruction, e.g. ‘Triletes spores’ in the example case study below). The fifth column ‘ProxyName’ contains the name of the taxon and columns two to four contain the taxonomic classification of that taxon as ‘Family’, ‘Genus’ and ‘Species’, respectively. For simplicity, a pre-formatted version of the PSE table with the names of all the taxa to study can be generated by crestr using the createPSE(list_of_taxa) function that generates a spreadsheet with the correct structure and with the Level and ProxyName columns automatically filled in.


Table 1 Example classification of four pollen taxa from the example case study, each one with a different level of taxonomic resolution. The last column ‘Taxonomic resolution’ is added here for illustrative purposes and should not included in the PSE table.

Level Family Genus Species ProxyName Taxonomic resolution
1 Asteraceae Asteraceae undiff. Family
2 Asteraceae Stoebe Stoebe-type Subfamily
2 Asteraceae Elytropappus Stoebe-type Subfamily
2 Asteraceae Artemisia Artemisia Genus level
3 Arecaceae Elaeis Elaeis guineensis Elaeis guineensis Species
4 Triletes spores To be excluded


When performing the classification with the crest.get_modern_data() function, crestr first classifies the taxa with the lowest taxonomical resolution (i.e. when Level is equal to one), and then increases the resolution Level by Level. In the example represented on Table 1, different taxonomic resolution levels are provided for different plant species belonging to the highly diverse Asteraceae family (the daisy family). To distribute all the Asteraceae species observed across the study area to their appropriate category, all the species are first classified as ‘Asteraceae undiff.’. In a second time, the classification of some of these Asteraceae species is refined when reaching the better resolved sub-groups (Stoebe-type and Artemisia at Level 2). At the end of the process, the ‘Asteraceae undiff.’ group only contains Asteraceae species that live in the study area but are not part of the genuses Stoebe, Elytropappus or Artemisia. The latter are categorised separately as Stoebe-type or Artemisia. Additional names can also be included to the PSE file to exclude species that are known to not be part of a group. For instance, this could have been used to simplify the climate response of the ‘Asteraceae undiff.’ group by excluding more species from it. This categorisation process can be time-consuming, as all the taxa must be classified in the same PSE table, and will often require many iterations to be optimised. The results of the different assignments are stored in the crestObj returned by the crest.get_modern_data() function and can be evaluated by checking rcnstrctn$misc$taxa_notes.


The alternative modern calibration dataset (distributions)

Users that prefer fitting proxy-climate relationships from their own calibration data should prepare a distributions dataset following the specific structure presented in Table 2. The first two columns should contain the species names (or unique identifiers) and the corresponding proxy name. If more than one species correspond to one taxon, the PDFs will be fitted in two steps, as explained here. The following two columns contain the coordinates of the species occurrence data. Finally, the last columns contain the climate values to be reconstructed. An optional column called weight can be added to distributions in fifth position (i.e. between the coordinates and the climate variables) if one wants to weight the different observations. For example, the (relative) abundance of taxa observed from modern data can be used when fitting the PDFs to give more importance to the observations where that abundance is highest.


Table 2 Template for the data frame. The weights column, here indicated with a ’*’, is optional and can be omitted or its values all set to 1 to assign the same weight to each observation. The number of rows the of table should correspond to the number of unique occurrences available.

Species name Taxon Name Longitude Latitude Weight* clim_1 … clim_n
Stoebe plumosa Stoebe-type 18.875 -34.375 20 15.8 … 711
Elytropappus rhinocerotis Stoebe-type 18.375 -33.625 32 16.9 … 477
… … … … … … … …
Elaeis guineensis Elaeis guineensis -4.375 10.875 4 27.4 … 1020


The selectedTaxa data frame

The last data frame that can be used to inform the reconstruction is a data frame of ones and zeros called selectedTaxa. This data frame has as many rows and columns as there are taxa and climate variables, respectively. Each entry should be either 1 or 0 and is used to indicate if the taxon should be used to reconstruct the climate variable (value = 1) or not (value = 0). If it is not provided, a default selectedTaxa data frame with all entries set to 1 is added to the crestObj. Users can modify this information at any point using the includeTaxa() and excludeTaxa() built-in functions. The crest.get_modern_data() function also modifies this data frame by setting the value to -1 when the PSE classification failed for a taxon or when the amount of data in the study area is insufficient to fit a reliable PDF.