knitr::opts_knit$set(root.dir = "//Users//abbey//Desktop//thesis//data//REPORT//data_summary")
mass_loss <- read.csv("mass_loss_compiled.csv")
#organize variables and define categorical variables
str(mass_loss)
## 'data.frame': 32 obs. of 21 variables:
## $ ML_harvest : int 1 1 1 1 1 1 1 1 1 1 ...
## $ sample : int 5710 5724 5729 5731 5736 5702 5704 5711 5726 5728 ...
## $ treatment : Factor w/ 2 levels "C","SC": 1 2 1 1 1 2 2 1 2 2 ...
## $ initial_wet_mass : num 0.838 0.483 0.769 0.668 1.179 ...
## $ shade : Factor w/ 5 levels "","dark","dark / med",..: 3 2 4 2 4 3 4 5 4 4 ...
## $ linear_model : Factor w/ 3 levels "","LM1","LM2": 2 2 3 2 3 2 3 3 3 3 ...
## $ initial_water_content : num 0.606 0.358 0.626 0.487 0.974 ...
## $ initial_dry_weight_calc : num 0.232 0.125 0.143 0.18 0.205 ...
## $ harvest_mass : num 0.914 0.525 1.02 0.906 1.5 ...
## $ final_dry_mass : num 0.183 0.102 0.122 0.218 0.18 ...
## $ mass_loss : num 0.0489 0.0226 0.0213 -0.0372 0.0247 ...
## $ X.mass_loss_from_initial: num 21.1 18.1 14.9 -20.6 12.1 ...
## $ mass_loss_from_OG : num 78.9 81.9 85.1 120.6 87.9 ...
## $ X : Factor w/ 3 levels "","CC_avg_ML",..: 3 2 1 1 1 1 1 1 1 1 ...
## $ X.1 : num 6.2 -2.98 NA NA NA ...
## $ X.2 : logi NA NA NA NA NA NA ...
## $ X.3 : Factor w/ 3 levels "","LM1","LM2": 1 2 3 1 1 1 1 1 1 1 ...
## $ X.4 : Factor w/ 3 levels "","dark","light": 1 2 3 1 1 1 1 1 1 1 ...
## $ X.5 : Factor w/ 4 levels "","-0.0276","0.02021",..: 4 3 2 1 1 1 1 1 1 1 ...
## $ X.6 : Factor w/ 4 levels "","0.69945","0.84974",..: 4 2 3 1 1 1 1 1 1 1 ...
## $ X.7 : Factor w/ 2 levels "","equation": 2 1 1 1 1 1 1 1 1 1 ...
mass_loss$ML_harvest=as.factor(mass_loss$ML_harvest)
#create boxplot describing percent mass loss from original dry mass
library(ggplot2)
mass_loss_plot <- ggplot(mass_loss, aes(x = ML_harvest, y = mass_loss_from_OG, fill = treatment))+
geom_boxplot()+
ggtitle("Boxplot of mean leaf litter mass loss by treatment") +
scale_y_continuous(name = "Mean percent mass loss",
breaks = seq(70, 145, 10),
limits=c(70, 145)) +
scale_x_discrete(name = "Harvest number")
mass_loss_plot
## Warning: Removed 11 rows containing non-finite values (stat_boxplot).
mass loss differs by treatment. the results are more straightforward for the snow cover treatment, in which there appears a decreasing trend for mass loss.
for current conditions, the results are not as straightforward. there is more variation within each time period, and mass loss was first shown and then mass gain was shown in the second harvest
this could be due to the variable conditions experienced. need to look closer at the literature for explanations + question: what could account for mass gain in litter decomposition experiments?
mass loss calculations are dependent on linear models created. 2 linear models created, one for ‘dark type’ leaves and one for ‘light type’ leaves. could potentially be erroneous categorization, include this in the discussion - I would have preferred to do the experimnet with only light type or dark type maple leaves to isolate this source of variation (particularly in water content and texture / assumed lignin concentration)
mass loss notes / to do