Title of the dataset:
Runoff, erosion and pesticide transport in a small catchment in Limburg - The Netherlands

Creators:
M. C. Commelin		ORCID: 0000-0001-7460-1915
Soil Physics and Land Management group, Wageningen University, 6708 WG Wageningen, The Netherlands 
contact: meindert.commelin@wur.nl

Contributors:
P. Peters
J. E. M. Baartman	ORCID: 0000-0001-6051-8619
P. Zomer		ORCID: 0000-0002-2498-963X
M. Riksen		ORCID: 0000-0002-3695-4362
V. Geissen


Related publication:
Pesticides are substantially transported in particulate phase, driven by land use, rainfall event and pesticide characteristics – a runoff and erosion study in a small agricultural catchment. M.C. Commelin et al. (2022).
DOI: 10.3389/fenvs.2022.830589 

Related software:
pesticide_transport_runoff_erosion: GitHub repository with the code to analyse this dataset.
DOI: 10.4121/19690840
URL: https://github.com/mcommelin/pesticide_transport_runoff_erosion

Description:
The main data in this dataset are observed rainfall runoff events. During this events runoff, erosion and related pesticide concentrations in the particulate and dissolved phase are measured. 
For each runoff event timeseries are available of measured discharge (1 minute interval), suspended sediment (5 minute interval) and pesticide concentrations in both phases. 
In addition, we collected data from field observations and interviews with land managers, data is available on pesticide applications and concentrations of pesticides in soil.

Keywords:
Rainfall-runoff
Pesticides
Pollution
Particulate phase
Erosion
Runoff

Spatial coverage:
38 ha catchment, Limburg, The Netherlands

Temporal coverage:
2018-05/2020-12

This dataset contains the following files:
overview_attributes.csv: file with explanation of all attributes in the raw data set.

Raw data files:
catchment_sample_analysis.csv: pH and OM analysis of samples on fields in the catchment
CR6.csv: the main timeseries data collected at the outlet of the catchment, including discharge and rainfall
CR1000_rain.csv: rainfall data collected at two substations in the catchment
CR1000_vwc.csv: soil moisture data collected at two substations in the catchment
Event_samples.csv: basic data for all samples collected during runoff events at the outlet
Event_samples_analysis.csv: pH and OM analysis of sample from the outlet
fields.csv: data on fields, including crops cultivated and area
KSAT.csv: results from hydraulic conductivity measurements in the catchment
LC_GLY.csv: raw LC-MS/MS data for glyphosate and ampa analysis
LC_multi.csv: raw LC-MS/MS data for multi residue analysis of pesticides
Pest_application.csv: application data of pesticides on fields in the catchment
Pest_describtion.csv: describtion of applied pesticides including aictive substances
Soil_samples.csv: data on all collected soil samples in the catchemnt, including field location
TDR_VWC.csv: manual collected soil moisture data in the catchment
tex_range_id.csv: texture range classes, to analysis texture_data.csv
texture_data.csv: texture analysis results for several sediment samples from the dataset
WB_data.csv: timeseries on water level and discharge at the catchment outlet as collected by the waterboard

Methods, materials and software:
For a describtion of the methods used to collect the main data see Commelin et al., (2022).

Methods not described in that paper:
-VWC collection
-KSAT measurement
-Air pressure to water level calculation - Parshall flume

Calculation of volumetric water content (VWC) based on raw data for CR1000_vwc time series:

The dataset contains time series with measurement of VWC and rainfall at two additional locations in the catchment. Due to miscommunication with landowners, the sensors where broken multiple times during tillage.
The dataset is therefor fragmented and not further used in the main publication. The VWC measurement are dan with Campbell Scientific CS616 time domain reflectrometers. Based on the response time the VWC is estimated.
The manual was retrieved from: https://s.campbellsci.com/documents/au/manuals/cs616.pdf.
The used calculation is given on page 12 of this manual, with an estimated +-3%: 
      VWC = -0.0663 + (-0.0063*PA_uS)+(0.0007*PA_uS^2)
with and PA_uS the respons period in micro seconds. The respons period is also provided in the data, so different calculations are possible.

KSAT (saturated hydraulic conductivity) sample collection and analysis:

For KSAT measurement, undisturbed soil rings from the topsoil (1 - 15 cm) were collected during a field campaign in 11-2019. The collection was done according the methods described in Stolte (1997)*.
The lab analysis was done conforming the NEN5789 protocol for Hydraulic conductivity measurement, in the Sediment and Hydrology laboratory at Wageningen University. 
Each ring was analysed in triplicate, with at least 10 minutes duration and 100 mL leachate water. KSAT was calculated with:

	KSAT = V / (|delta H| * delta t * A)
with 
	|delta H| = (z_top - z_out) / L and q = V / (delta t * A)

In which V is the volume of water that was collected in 'delta t' time, L the height of the sample and A the area of the sample.

*Stolte, J. (1997). Manual for soil physical measurements: version 3'. DLO Winand Starting Centre.

Calculation of water level based on pressure sensors (Hb) in Parshall flume:

The downstream water level in the Parshall flume, was measured with two pressure sensors (STS ATM sensor). The two sensors were installed on a fixed height from each other (dH).
The difference in pressure reading thus is the results of the water and air pressure difference between the 2 sensors. The response in milivolt was converted to air pressure in pascal.
This is used to calculate the layer of water on top of the lower pressures sensor.

The calculation to pascal was done with: 
    P=(950+(STS_press(1)/5000*600))*100
with STS_press(i) the millivolt reading of the sensor. From this the waterlevel was calculated with:
    PBair = PA - (dH * PaM)
with PBair, the air pressure part of the lower sensor reading, PA the air pressure of the upper sensor, dH the height difference in meter and PaM the pascal per meter height (11.3 Pa/m).
From here the level of water above the s=lower sensors is calculated:
    STS_level = (PB - PBair)/(ROw*g)
with ROw the density of water (997 g/L) and g the gravitation (9.80665 m/s2), the final water level was found by correcting for the difference between the barometer and the inlet in the flume. Which was found by calibratrion.
   Wat_Level = STS_level - dHc - cal
with dHc, the measured difference (+- 3mm) and cal a parameter for calibration based on water level observations in the flume.


This dataset is published under the CC BY (Attribution) license.
This license allows reusers to distribute, remix, adapt, and build upon the material in any medium or format, so long as attribution is given to the creator.