Data from: How does soil organic matter affect potato productivity on sandy soil?
Data files
Jun 20, 2025 version files 17 KB
-
README.md
4.28 KB
-
SOM_DATA_2Aug2023.csv
12.72 KB
Abstract
While many studies note the positive effects of soil organic matter (SOM) on crop yields, there is limited quantitative information on the influence of increased SOM on potato productivity in sandy soil. This study estimated the impact of varying SOM on potato productivity in sandy soils and explored whether nitrogen (N) mineralization served as a primary mediator. Soil from nine fields in Wisconsin (SOM range of 1.1 to 3.8 %) was collected for a greenhouse study. Both NH4-N and NO3-N extracted from ion strips and potentially mineralizable nitrogen (PMN) were used as the proxies for N mineralization. Linear mixed effect models indicated that fresh matter whole biomass and dry matter vine biomass were 0.45 and 0.54 times greater at 3.8 % SOM compared to 1.1 % SOM at an optimal soil pH of 5.2, respectively. Similarly, total N uptake in the whole and vine biomass was 0.51 and 1.0 times higher at 3.8 % SOM than 1.1 % SOM, respectively. While PMN demonstrated a positive correlation with SOM, it only partially mediated the effect of SOM on productivity, specifically in N uptake in the vines. However, for most productivity measures, including PMN with SOM in the models did not substantially reduce the estimated SOM effect on productivity, indicating that SOM affected productivity mostly through mechanisms other than N acquisition by plants available through mineralization of OM. The study underscored the complex interplay between SOM and potato productivity, urging further research into the multifaceted roles of SOM in sandy soils.
https://doi.org/10.5061/dryad.wstqjq2t9
This dataset investigates the impact of Soil Organic Matter (SOM) on potato productivity in sandy soil. The data is derived from an experimental study that includes various treatments, replications, and measured parameters related to soil properties and potato growth. The dataset provides insights into the relationships between soil characteristics and potato productivity.
Description of the data and file structure
The dataset consists of several columns representing different aspects of the experimental setup, soil properties, and potato growth. Some key columns include:
Treatments: Different experimental treatments applied.
N1 = 0 nitrogen
S1, 2, 3 = Replications 1, 2, 3
FF, NF, W, L, RE, TN, DH, GS, NW= Nine different abbreviated fieldnames
Reps: Replications of each treatment.
Fieldnames: Descriptive field names.
Farms: Farms associated with the study.
(Other columns representing various soil properties, rates, and measurements related to potato growth)
The data is organized in a tabular format with each row representing a specific combination of treatment, replication, and fields.
The data is contained within a single file, and relationships between records are established through shared identifiers such as treatment codes and replication numbers.
Abbreviations
SOM = Soil organic matter
perc= percent
ppm = parts per million
pp = pre planting
mgperkg = milligram per kilogram
mgperL = milligram per liter
g = gram
num = numbers
Lab = lab measured
TN= Total nitrogen
up = uptake
TN_up_vines = total nitrogen uptake in the vines
TN_perc_vines = total nitrogen content in the vines (measured as percentage)
CEC= Cation exchange capacity
Ca, Mg, P, K, Na, Cu, Zn, Fe, Mn = Calcium, Magnesium, Phosphorus, Potassium, Sodium, Copper, Zinc, Iron, Manganese
NH4, NO3 = ammonium, nitrate
Mar, Apr, May, Jun = Abbreviations for study months
Sharing/Access information
Data was derived from the experimental study.
Code/Software
This script used is R analysis script for a dataset related to soil properties and plant biomass.
Break down the code into sections:
Data Import and Linear Mixed Effects Models (LME):
SOM <- read.csv (“SOM_DATA_2Aug2023.csv”, header = T)
labels (SOM)
library(lme4)
library(lmerTest)
whole <- lmer(fresh_whole_g ~ SOM_perc + Soil_pH + (1|Fieldnames), data = SOM)
summary(whole)
# Similar models for DMvine, TNupwhole, and TNupvines
This part reads a CSV file, defines a data frame (SOM), and fits linear mixed-effects models (lmer) for different response variables.
Data Visualization (Step 1):
# Several ggplot2 visualizations for different response variables and their relationships with predictors
# Each plot includes a linear regression line, points, and additional visual elements
This section creates various plots using ggplot2 to visualize the relationships between different response variables and predictors.
Potentially Mineralizable N and In-situ (seasonal) analysis (Step 2 and Step 3):
# Models and visualizations related to potentially mineralizable N, as well as seasonal analysis
# These include linear models, ANOVA, and ggplot2 visualizations
Here, the script models and visualizes the potentially mineralizable N and performs a seasonal analysis using linear mixed-effects models and ANOVA.
Final Model (Step 4):
# A final linear mixed-effects model and its visualization using ggplot2
final <- lmer(TN_up_vines ~ SoilNH3_mgperkg_lab + SOM_perc + Soil_pH + (1|Fieldnames), data = SOM)
summary(final)
# Visualization using ggplot2 with a color gradient
This section fits a final linear mixed-effects model and visualizes the relationship between the response variable and predictors.
Notes:
The script heavily relies on the lme4 and ggplot2 packages for modeling and visualization.
The script follows a structured approach with labeled steps.
The results are summarized and visually presented at various steps.
This is a comprehensive script that covers data import, modeling, visualization, and analysis for the given dataset.