Nitrate leaching experiment with time-series data

Dataset

Engstrom et al. 2011 reported results from an experimental field trial in Sweden which assessed the effects of different fertilization levels and crop rotations on nitrate leaching. There were two experiments with nine treatments over two years, resulting in 36 Cycles. Leaching data were provided with monthly frequency. Data were presented in tables and the text.

Conversion steps

We largely followed the same steps as the nitrate leaching meta-analysis example, with the following additional steps:

• Data for many Measurements, Inputs, and Emissions were provided as arrays where each value represented the date they occurred on. For upload to HESTIA, arrays can be created in Excel or csv files by separating each value with a semicolon (e.g., 0.443;23.461). Arrays of associated dates are added in the dates field using the ISO 8601 format (e.g., 2004-09;2004-10, which represent September 2004 and October 2004 respectively). Missing values in arrays are represented using either a dash or a blank (e.g., 4;-;8, 4;;8).
• Long term rainfall and temperature data were provided as the total or average for each month averaged over the period 1961-1990. To represent these data, the startDate and endDate of the Measurements were specified as 1961 and 1990 respectively. The monthly rainfall and temperature data were added using an array. Arrays of associated dates were added in the dates field using the ISO 8601:2000 format for dates without a year (e.g., --01;--02;... , which represent January and February respectively).
• The values in some arrays are additive (e.g., monthly rainfall sums to annual rainfall), whereas others are not (e.g., monthly temperature averages to annual average temperature). The nitrogen uptake by a plant is an additive over time, which means that the value for each date must be the incremental amount of uptake compared to the previous date. The uptake data in the study were already in this format and were entered directly.

Validation

Soft validation errors were raised for the twelve rapeseed Cycles recommending that the crop yield was provided, however these data were not available in the study and this error was ignored. Soft validation errors were also raised where soil texture ("sandy loam") as provided without a depth interval and based on the article a depth interval of 0-30cm was assigned, and this was corrected.

Downloads

Note: these data are part of a larger dataset.

• Dataset in HESTIA format (Excel)
• Dataset in HESTIA format (JSON)
• Data in the Explorer