We have continued to solicit feedback from the broader community of FRED users; the updates that we implemented between FRED 1.0 and FRED 2.0 can be found below. Because FRED 2.0 improves upon FRED 1.0, and also corrects some minor mistakes, we recommend that users download FRED 2.0. However, FRED 1.0 is also still available.
Compared to FRED 1.0, FRED 2.0 features a total increase of approximately 35,000 root trait observations, collected from over 400 additional sources.
We have added 7950 new rows to FRED since version 1. However, some of the original rows of data from FRED 1.0 have been removed due to redundancy or poor data; for this reason, the highest Row ID (35475) exceeds the total number of rows in the dataset (35211). Data are unique to FRED 2.0 for all rows where “2” appears in column F00001 (Version_FRED version number).
Rows 35446 to 35459 are attributed to FRED 1.0 in column F00001, although these rows are not present in FRED 1.0. This is because the data in these rows were moved from their original position (listed as covariate with a less-specific pool of roots) to new columns for the sake of greater clarity.
FRED 1.0 contained some erroneous or incomplete data which we corrected before releasing FRED 2.0. Some data which did not stand out during the quality-checking process for FRED 1.0 became more apparent when compared to a larger data pool for FRED 2.0. Other erroneous or incomplete data were reported by users or noticed in the process of entering new data. Users are encouraged to reach out to us through the contact page to report suspicious or erroneous data.
Table 5 in the User Guidance Document lists corrections made between FRED 1.0 and FRED 2.0.
Our focus during the expansion from FRED 1.0 to FRED 2.0 was to fill existing columns rather than add columns for new traits. However, we did add new columns if we deemed them particularly useful or if new statistical metrics, such as n, SE, or SD, were available for existing values.
Table 6 in the User Guidance Document lists new columns present in FRED 2.0 that were not present in FRED 1.0.
We most often removed columns because we considered them redundant with other columns, as with columns F00015 (Plant taxonomy_Group, redundant with F01291, Plant taxonomy_Group_TPL) and F00190 (Root tips per root mass, redundant with F00192, Specific root tip abundance (SRTA)), or because they were empty or underpopulated, as with columns F00098 (n_Root aerenchyma porosity fraction) and F00363 (Root lipid content per root C content).
Table 7 in the User Guidance Document lists columns which were present in FRED 1.0 but removed from FRED 2.0.
We renamed columns for further clarity, as in column F00582 (Root turnover_Annual root system replacement, formerly Root turnover_Mass per year), or to reflect a slight change or correction in the column’s function or content, as in columns F00980 (Soil pH_Water, formerly Soil pH) and F001156 (Notes_In situ, pot, or hydroponic; formerly Notes_In situ or pot).
Table 8 in the User Guidance Document lists columns which were present in FRED 1.0 but renamed for FRED 2.0.