Chapter 1- PREDICTING NATURAL ARTHROPOD DENSITY THROUGH GENERALIST HERBIVORE ASSAYS
Tyler Zarubin
2024-02-15
- 1- Load plant/arthropod trait data & combine tables, wrangle data
- 2- Perform
Linear Mixed Effects Regression analysis
- 2.1 Visualize data, establish normality
- 2.2.1 log Total Abundance ~ Spodoptera percent survival
- 2.2.2 log Herbivore Abundance ~ Spodoptera percent survival
- 2.2.3 log Predator Abundance ~ Spodoptera percent survival
- 2.2.4 log Omnivore abundance ~ Spodoptera percent survival
- 2.2.5 log Detritivore abundance ~ Spodoptera percent survival
- Kailen’s to-do list
knitr::opts_chunk$set(echo = TRUE, out.width = '100%')1- Load plant/arthropod trait data & combine tables, wrangle data
This study seeks to address how strongly laboratory leaf consumption assays with a generalist herbivore predict naturally occurring arthropod counts, mass and densities on native plants. The metadata used for our analyses was collected each year in April-May from 2021-2023 in a common garden at Concordia University Irvine planted with woody shrubs native to Southern California. The common garden is made up of 20 separate plots measuring 6m x 5m, each with 1 individual plant per species. Plants were grown from seed in a greenhouse and transplanted after 1 year. Data collected, abbreviations used and variable type assigned for the metadata can be found in the table below. NOTE: Log values of arthropod counts and mass (not shown in the table below) were also calculated if necessary to establish normality of data.
| Data | Description | Column Name | Variable Type |
|---|---|---|---|
| Plant # | Individual plant | Plant | Factor |
| Plant Species | Plant Species | Species | Factor |
| Watering Treatment | Water added to plots ranging from Ambient (1) to High (6) | Watering_Treatment | Factor |
| Plot | Individual Plot | Plot | Factor |
| Year | Year data collected | Year | Factor |
| Spodoptera percent survival | % Spodoptera surviving at end of trial | Spod_perc_survival | Numeric |
| Spodoptera surviving | Total number of Spodoptera surviving at end of trial | Spod_surv_total | Numeric |
| Average Spodoptera mass (g) | Average mass of surviving Spodoptera | Avg_spod_mass_g | Numeric |
| Total arthropod count | Total number of arthropods | Abundance_Total | Numeric |
| Herbivore count | Total number of herbivores | Abundance_Herb | Numeric |
| Predator count | Total number of Predators | Abundance_Pred | Numeric |
| Omnivore count | Total number of Omnivores | Abundance_Omni | Numeric |
| Detritivore count | Total number of Detritivores | Abundance_Detr | Numeric |
| Palynivore count | Total number of Palynivores | Abundance_Paly | Numeric |
| Total arthropod mass (mg) | Combined mass of all arthropods | Arth_Mass_Total_mg | Numeric |
| Herbivore mass (mg) | Combined mass of all herbivbores | Mass_Herb_mg | Numeric |
| Predator mass (mg) | Combined mass of all predators | Mass_Pred_mg | Numeric |
| Omnivore mass (mg) | Combined mass of all omnivores | Mass_Omni_mg | Numeric |
| Detritivore mass (mg) | Combined mass of all detritivores | Mass_Detr_mg | Numeric |
| Palynivore mass (mg) | Combined mass of all palynivores | Mass_Paly_mg | Numeric |
| Estimated dry plant biomass (g) | Estimated dry biomass of aboveground portion of plant | Dry_Plant_Biomass_g | Numeric |
| Estimated plant biomass sampled | Estimated dry biomass of plant sampled during arthropod collection | Dry_Biomass_Sampled_g | Numeric |
Spodoptera data was collected from a 10-day feeding assay where Spodoptera exigua, a generalist herbivore, were hatched from eggs and immediately fed leaves from each individual plant. At the end of 10 days, the number of surviving larvae and final mass of each larvae were recorded. See You, An, and Li (2020) for details. Plant and arthropod data were collected from each surviving plant from 2021-2023. Plant estimated biomass was collected using a branch-biomass estimation method as previously described in Nell and Mooney (2019). Arthropods were collected from plants using a vaccum method as previously described in Pratt et al. (2017), sorted from plant material and identified to feeding guild. For 2023 arthropod samples, lengths of arthropods were measured and used to estimate arthropod dry biomass. See Rogers, Hinds, and Buschbom (1976) for details.
1.1 Load packages LMER analysis
1.2 Wrangle data
All data frames from arthropod, plant biomass and Spodoptera bioassays from 2021-2023 were combined into a single dataframe. In addition, we introduced log transformations for arthropod counts and arthropod mass columns. NOTE: for plant samples where 0 arthropods were collected, we manually entered “0” after log transformation to avoid errors when calculating the log of 0.
2- Perform Linear Mixed Effects Regression analysis
I will be using the general formula
glmmTMB(arthropod ~ bioassay * species * year + (1|Watering_treatment) + (1|Plot) + (1|Plant), data = total_plant)
for all analyses with arthropod being count or mass of
numbers within each feeding guild or total arthropods on a plant and
bioassay being either the average spodoptera mass or
number/percent of spodoptera surviving at the end of the trial. Plot,
Watering Treatment and Plant will be added as random factors to the
model.
These results will help to determine the strength by which feeding assay data is predictive of naturally occurring arthropod communities.
2.1 Visualize data, establish normality
Plot regressions of Arthropod abundance ~ Spodoptera across species. Visually assess the distribution of data points as well as slopes. Determine if arthropod counts are Poisson-distributed, log transform to establish normality if necessary.
## Create new datasets, one with Avg Spodoptera Mass and another with individual spodoptera mass
# Scale Spodoptera mass to standard deviations
total_plant2 = total_plant %>%
select(Abundance_Total, Abundance_Herb, Abundance_Omni, Abundance_Pred, Avg_spod_mass_g, Spod_perc_survival, Species,Year, Plot, Watering_Treatment, Plant) %>%
mutate(Avg_spod_mass_g_scale = scale(Avg_spod_mass_g))
total_2 = total %>%
select(Abundance_Total, Abundance_Herb, Abundance_Omni, Abundance_Pred, Spod_mass, Spod_perc_survival, Species,Year, Plot, Watering_Treatment, Plant) %>%
mutate(Spod_mass_scale = scale(Spod_mass))
## Plot Total abundance ~ Avg. Spodoptera mass
ggplot(total_plant2,aes(y = log(Abundance_Total + 1), x = Avg_spod_mass_g_scale))+
geom_point()+
geom_smooth(method = "lm")+
facet_wrap(~Species) +
ggtitle("Total abundance ~ Avg. Spodoptera mass")+
xlab("Avg. Spodoptera mass (SD)")+
ylab("log Total Abundance")## `geom_smooth()` using formula = 'y ~ x'
## Plot Total abundance ~ Spodoptera mass
ggplot(total_2,aes(y = log(Abundance_Total + 1), x = Spod_mass_scale))+
geom_point()+
geom_smooth(method = "lm")+
facet_wrap(~Species) +
ggtitle("Total abundance ~ Spodoptera mass")+
xlab("Spodoptera mass (SD)")+
ylab("log Total Abundance")## `geom_smooth()` using formula = 'y ~ x'
## Plot Herbivore abundance ~ Avg. Spodoptera mass
ggplot(total_plant2,aes(y = log(Abundance_Herb + 1), x = Avg_spod_mass_g_scale))+
geom_point()+
geom_smooth(method = "lm")+
facet_wrap(~Species)+
ggtitle("Herbivore abundance ~ Avg. Spodoptera mass")+
xlab("Avg. Spodoptera mass (SD)")+
ylab("log Herbivore abundance")## `geom_smooth()` using formula = 'y ~ x'
## Plot Herbivore abundance ~ Spodoptera mass
ggplot(total_2,aes(y = log(Abundance_Herb + 1), x = Spod_mass_scale))+
geom_point()+
geom_smooth(method = "lm")+
facet_wrap(~Species)+
ggtitle("Herbivore abundance ~ Spodoptera mass")+
xlab("Spodoptera mass (SD)")+
ylab("log Herbivore abundance")## `geom_smooth()` using formula = 'y ~ x'
## Plot Total abundance ~ Spodoptera Survival
ggplot(total_plant2,aes(y = log(Abundance_Total + 1), x = Spod_perc_survival))+
geom_point()+
geom_smooth(method = "lm")+
facet_wrap(~Species) +
ggtitle("Total abundance ~ Spodoptera survival (%)")+
xlab("Spodoptera percent survival")+
ylab("log Total Abundance")## `geom_smooth()` using formula = 'y ~ x'
## Plot Herbivore abundance ~ Spodoptera Survival
ggplot(total_plant2,aes(y = log(Abundance_Herb + 1), x = Spod_perc_survival))+
geom_point()+
geom_smooth(method = "lm")+
facet_wrap(~Species)+
ggtitle("Herbivore abundance ~ Spodoptera survival (%)")+
xlab("Spodoptera percent survival")+
ylab("log Herbivore abundance")## `geom_smooth()` using formula = 'y ~ x'
Based on the plots, I can make the following conclusions: - Arthropod counts ~ Avg. Spodoptera Mass - There is very little variance (<1 SD) in average spodoptera mass for most species. This presents a problem in that very small changes in spodoptera mass have an outsized effect on the model resulting in extremely large confidence intervals. - For those species that do have considerable variance (2 SD or more), it seems like a neutral/ slight negative effect of increasing spodoptera mass on total counts. - Regardless, I do not think that the predictor variable in this case is appropriate to draw conclusions from. - Herbivore counts show similar patterns - Arthropod counts ~ Spodoptera Mass - Here I used the individual measurements of spodoptera masses instead of the average. The thought was that this would provide potentially greater variance among the predictor variable and allow for stronger confidence in the model. The results were disappointing in that the same 3 species (ACMGLA, ERIFAS, and SALAPI) show variance among the predictor >2 Standard deviations. - Herbivore abundance shows no strong difference from total abundance - Arthropod counts ~ Spodoptera percent survival - The variance seen here, both for total arthropods and herbivores only is wide and, I think, offers a better predictor variable to run my model off of.
TAKEAWAY: I will ingnore spodoptera biomass as a predictor and instead focus solely on percent spodoptera survival as my fixed effect
2.2.1 log Total Abundance ~ Spodoptera percent survival
This analysis will determine the strength of the relationship between the percent survival of Spodoptera and total arthropod counts from each plant.
## Run glmmTMB model for Total abundance ~ Spodoptera percent survival. Family = poisson.
arthcount_surv <- glmmTMB(Abundance_Total ~ Spod_perc_survival * Species + Year + (1|Plot) + (1|Watering_Treatment) + (1|Plant), data = total_plant, family = poisson())
##Test dispersion with Dharma package.
testDispersion(arthcount_surv)
##
## DHARMa nonparametric dispersion test via sd of residuals fitted vs.
## simulated
##
## data: simulationOutput
## dispersion = 0.67376, p-value = 0.68
## alternative hypothesis: two.sided## results indicate the data is not over- or under-dispersed and we can stick with poisson and not have to switch to negative binomial.
Anova(arthcount_surv, type = 3)| Chisq | Df | Pr(>Chisq) |
|---|---|---|
| 694 | 1 | 6.18e-153 |
| 1.31e+03 | 1 | 3.68e-286 |
| 296 | 13 | 2.07e-55 |
| 5.52e+03 | 2 | 0 |
| 2.35e+03 | 13 | 0 |
summary(arthcount_surv)## Family: poisson ( log )
## Formula: Abundance_Total ~ Spod_perc_survival * Species + Year + (1 |
## Plot) + (1 | Watering_Treatment) + (1 | Plant)
## Data: total_plant
##
## AIC BIC logLik deviance df.resid
## 25181.6 25327.0 -12557.8 25115.6 573
##
## Random effects:
##
## Conditional model:
## Groups Name Variance Std.Dev.
## Plot (Intercept) 2.899e-08 0.0001703
## Watering_Treatment (Intercept) 1.673e-03 0.0409079
## Plant (Intercept) 7.398e-01 0.8601439
## Number of obs: 606, groups: Plot, 20; Watering_Treatment, 6; Plant, 246
##
## Conditional model:
## Estimate Std. Error
## (Intercept) 6.06188 0.23011
## Spod_perc_survival -3.73135 0.10322
## SpeciesArtemisia_californica -0.69232 0.29948
## SpeciesDiplacus_aurantiacus -2.17611 0.30094
## SpeciesEncelia_californica -1.55796 0.29997
## SpeciesEriogonum_fasciculatum -2.15250 0.31819
## SpeciesGrindelia_camporum -0.98167 0.31340
## SpeciesIsocoma_menziesii -1.28848 0.31505
## SpeciesMalacothamnus_fasciculatus -0.50207 0.31435
## SpeciesMalacothrix_saxatilis -2.52274 0.32203
## SpeciesMirabilis_laevis -1.94665 0.30064
## SpeciesSalvia_apiana -0.74223 0.31044
## SpeciesSalvia_mellifera -1.33430 0.30703
## SpeciesSisyrinchium_bellum -3.36171 0.31797
## SpeciesStachys_ajugoides_var_rigida -3.35587 0.30912
## Year2022 -0.93215 0.01264
## Year2023 -0.38349 0.01137
## Spod_perc_survival:SpeciesArtemisia_californica 2.26703 0.12059
## Spod_perc_survival:SpeciesDiplacus_aurantiacus 4.73806 0.12981
## Spod_perc_survival:SpeciesEncelia_californica 1.52150 0.22847
## Spod_perc_survival:SpeciesEriogonum_fasciculatum 4.64286 0.12173
## Spod_perc_survival:SpeciesGrindelia_camporum 2.18963 0.15737
## Spod_perc_survival:SpeciesIsocoma_menziesii 2.92747 0.14667
## Spod_perc_survival:SpeciesMalacothamnus_fasciculatus 4.13995 0.12541
## Spod_perc_survival:SpeciesMalacothrix_saxatilis 3.58163 0.17486
## Spod_perc_survival:SpeciesMirabilis_laevis 3.99961 0.19002
## Spod_perc_survival:SpeciesSalvia_apiana 3.41274 0.11196
## Spod_perc_survival:SpeciesSalvia_mellifera 3.25567 0.13283
## Spod_perc_survival:SpeciesSisyrinchium_bellum 3.22019 0.25708
## Spod_perc_survival:SpeciesStachys_ajugoides_var_rigida 2.64716 0.93060
## z value Pr(>|z|)
## (Intercept) 26.34 < 2e-16 ***
## Spod_perc_survival -36.15 < 2e-16 ***
## SpeciesArtemisia_californica -2.31 0.02079 *
## SpeciesDiplacus_aurantiacus -7.23 4.79e-13 ***
## SpeciesEncelia_californica -5.19 2.06e-07 ***
## SpeciesEriogonum_fasciculatum -6.76 1.33e-11 ***
## SpeciesGrindelia_camporum -3.13 0.00173 **
## SpeciesIsocoma_menziesii -4.09 4.32e-05 ***
## SpeciesMalacothamnus_fasciculatus -1.60 0.11022
## SpeciesMalacothrix_saxatilis -7.83 4.73e-15 ***
## SpeciesMirabilis_laevis -6.48 9.48e-11 ***
## SpeciesSalvia_apiana -2.39 0.01681 *
## SpeciesSalvia_mellifera -4.35 1.39e-05 ***
## SpeciesSisyrinchium_bellum -10.57 < 2e-16 ***
## SpeciesStachys_ajugoides_var_rigida -10.86 < 2e-16 ***
## Year2022 -73.77 < 2e-16 ***
## Year2023 -33.73 < 2e-16 ***
## Spod_perc_survival:SpeciesArtemisia_californica 18.80 < 2e-16 ***
## Spod_perc_survival:SpeciesDiplacus_aurantiacus 36.50 < 2e-16 ***
## Spod_perc_survival:SpeciesEncelia_californica 6.66 2.75e-11 ***
## Spod_perc_survival:SpeciesEriogonum_fasciculatum 38.14 < 2e-16 ***
## Spod_perc_survival:SpeciesGrindelia_camporum 13.91 < 2e-16 ***
## Spod_perc_survival:SpeciesIsocoma_menziesii 19.96 < 2e-16 ***
## Spod_perc_survival:SpeciesMalacothamnus_fasciculatus 33.01 < 2e-16 ***
## Spod_perc_survival:SpeciesMalacothrix_saxatilis 20.48 < 2e-16 ***
## Spod_perc_survival:SpeciesMirabilis_laevis 21.05 < 2e-16 ***
## Spod_perc_survival:SpeciesSalvia_apiana 30.48 < 2e-16 ***
## Spod_perc_survival:SpeciesSalvia_mellifera 24.51 < 2e-16 ***
## Spod_perc_survival:SpeciesSisyrinchium_bellum 12.53 < 2e-16 ***
## Spod_perc_survival:SpeciesStachys_ajugoides_var_rigida 2.84 0.00445 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1## look at plot of model
plot_model(arthcount_surv, type = "pred", terms = c("Spod_perc_survival", "Species"), se - T,ci.lvl = .5)+
ggtitle("Predicted total arthropod counts as a function of spodoptera survival")## Warning in RColorBrewer::brewer.pal(n, pal): n too large, allowed maximum for palette Set1 is 9
## Returning the palette you asked for with that many colors
TAKEAWAY Percent survival of spodoptera is a strong predictor overall of total arthropod counts with most species suggesting a negative relationship between spodoptera survival and total arthropod abundance. Which doesn’t make sense!!! Shouldn’t more spodoptera surviving to the end of the trial indicate a higher quality plant and, consequently, a plant that should contain more arthropods??
When I took away ACMGLA and STAAJU (for reasons stated above), the model still showed a strongly negative relationship between spodoptera survival and total arthropod count. This was the same with Species alone. The interactive effect was mostly positive indicating that this interaction produced a more positive relationship than that suggested by either fixed effect alone.
As for the random effects, “plot” and “watering treatment” had very little effect on the overall results and I will consider dropping them. “plant”, however, had a large effect on the overall model with a variance of .49 and a standard deviation of .7. This seems high so I think it is important to keep plant as a random effect in the model.
2.2.2 log Herbivore Abundance ~ Spodoptera percent survival
This analysis will determine the strength of the relationship between the percent survival of Spodoptera and herbivore counts from each plant.
## Run glmmTMB model for Herbivore abundance ~ Spodoptera percent survival. Family = poisson.
herbcount_surv <- glmmTMB(Abundance_Herb ~ Spod_perc_survival * Species + Year + (1|Plot) + (1|Watering_Treatment) + (1|Plant), data = total_plant, family = poisson())
##Test dispersion with Dharma package.
testDispersion(herbcount_surv)
##
## DHARMa nonparametric dispersion test via sd of residuals fitted vs.
## simulated
##
## data: simulationOutput
## dispersion = 0.3509, p-value = 0.28
## alternative hypothesis: two.sided## results indicate the data is not over- or uner-dispersed and we can stick with poisson and not have to switch to negative binomial.
Anova(herbcount_surv, type = 3)| Chisq | Df | Pr(>Chisq) |
|---|---|---|
| 514 | 1 | 6.93e-114 |
| 1.82e+03 | 1 | 0 |
| 292 | 13 | 1.01e-54 |
| 7.66e+03 | 2 | 0 |
| 2.6e+03 | 13 | 0 |
summary(herbcount_surv)## Family: poisson ( log )
## Formula:
## Abundance_Herb ~ Spod_perc_survival * Species + Year + (1 | Plot) +
## (1 | Watering_Treatment) + (1 | Plant)
## Data: total_plant
##
## AIC BIC logLik deviance df.resid
## 21333.5 21478.9 -10633.7 21267.5 573
##
## Random effects:
##
## Conditional model:
## Groups Name Variance Std.Dev.
## Plot (Intercept) 4.067e-08 2.017e-04
## Watering_Treatment (Intercept) 4.110e-09 6.411e-05
## Plant (Intercept) 1.143e+00 1.069e+00
## Number of obs: 606, groups: Plot, 20; Watering_Treatment, 6; Plant, 246
##
## Conditional model:
## Estimate Std. Error
## (Intercept) 6.43058 0.28353
## Spod_perc_survival -5.10315 0.11973
## SpeciesArtemisia_californica -1.50763 0.37141
## SpeciesDiplacus_aurantiacus -3.46789 0.37454
## SpeciesEncelia_californica -2.40977 0.37231
## SpeciesEriogonum_fasciculatum -3.24932 0.39447
## SpeciesGrindelia_camporum -1.51502 0.38834
## SpeciesIsocoma_menziesii -1.88050 0.39121
## SpeciesMalacothamnus_fasciculatus -1.04421 0.39012
## SpeciesMalacothrix_saxatilis -3.72735 0.40229
## SpeciesMirabilis_laevis -2.97284 0.37384
## SpeciesSalvia_apiana -1.29731 0.38565
## SpeciesSalvia_mellifera -1.85795 0.38013
## SpeciesSisyrinchium_bellum -4.33256 0.39738
## SpeciesStachys_ajugoides_var_rigida -4.03627 0.38315
## Year2022 -1.08599 0.01466
## Year2023 -1.00499 0.01535
## Spod_perc_survival:SpeciesArtemisia_californica 4.41055 0.14502
## Spod_perc_survival:SpeciesDiplacus_aurantiacus 6.91678 0.16813
## Spod_perc_survival:SpeciesEncelia_californica 1.78953 0.33105
## Spod_perc_survival:SpeciesEriogonum_fasciculatum 6.43344 0.15360
## Spod_perc_survival:SpeciesGrindelia_camporum 3.37444 0.18714
## Spod_perc_survival:SpeciesIsocoma_menziesii 3.53126 0.21300
## Spod_perc_survival:SpeciesMalacothamnus_fasciculatus 5.54845 0.14267
## Spod_perc_survival:SpeciesMalacothrix_saxatilis 5.50593 0.21449
## Spod_perc_survival:SpeciesMirabilis_laevis 5.45587 0.28006
## Spod_perc_survival:SpeciesSalvia_apiana 4.83499 0.13037
## Spod_perc_survival:SpeciesSalvia_mellifera 4.69631 0.15655
## Spod_perc_survival:SpeciesSisyrinchium_bellum 3.72058 0.36736
## Spod_perc_survival:SpeciesStachys_ajugoides_var_rigida 2.89083 1.29330
## z value Pr(>|z|)
## (Intercept) 22.68 < 2e-16 ***
## Spod_perc_survival -42.62 < 2e-16 ***
## SpeciesArtemisia_californica -4.06 4.92e-05 ***
## SpeciesDiplacus_aurantiacus -9.26 < 2e-16 ***
## SpeciesEncelia_californica -6.47 9.64e-11 ***
## SpeciesEriogonum_fasciculatum -8.24 < 2e-16 ***
## SpeciesGrindelia_camporum -3.90 9.57e-05 ***
## SpeciesIsocoma_menziesii -4.81 1.53e-06 ***
## SpeciesMalacothamnus_fasciculatus -2.68 0.007437 **
## SpeciesMalacothrix_saxatilis -9.27 < 2e-16 ***
## SpeciesMirabilis_laevis -7.95 1.83e-15 ***
## SpeciesSalvia_apiana -3.36 0.000768 ***
## SpeciesSalvia_mellifera -4.89 1.02e-06 ***
## SpeciesSisyrinchium_bellum -10.90 < 2e-16 ***
## SpeciesStachys_ajugoides_var_rigida -10.53 < 2e-16 ***
## Year2022 -74.06 < 2e-16 ***
## Year2023 -65.49 < 2e-16 ***
## Spod_perc_survival:SpeciesArtemisia_californica 30.41 < 2e-16 ***
## Spod_perc_survival:SpeciesDiplacus_aurantiacus 41.14 < 2e-16 ***
## Spod_perc_survival:SpeciesEncelia_californica 5.41 6.46e-08 ***
## Spod_perc_survival:SpeciesEriogonum_fasciculatum 41.88 < 2e-16 ***
## Spod_perc_survival:SpeciesGrindelia_camporum 18.03 < 2e-16 ***
## Spod_perc_survival:SpeciesIsocoma_menziesii 16.58 < 2e-16 ***
## Spod_perc_survival:SpeciesMalacothamnus_fasciculatus 38.89 < 2e-16 ***
## Spod_perc_survival:SpeciesMalacothrix_saxatilis 25.67 < 2e-16 ***
## Spod_perc_survival:SpeciesMirabilis_laevis 19.48 < 2e-16 ***
## Spod_perc_survival:SpeciesSalvia_apiana 37.09 < 2e-16 ***
## Spod_perc_survival:SpeciesSalvia_mellifera 30.00 < 2e-16 ***
## Spod_perc_survival:SpeciesSisyrinchium_bellum 10.13 < 2e-16 ***
## Spod_perc_survival:SpeciesStachys_ajugoides_var_rigida 2.24 0.025402 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1## look at plot of model
plot_model(herbcount_surv, type = "pred", terms = c("Spod_perc_survival", "Species"), se - T,ci.lvl = .5)+
ggtitle("Predicted total herbivore counts as a function of spodoptera survival")## Warning in RColorBrewer::brewer.pal(n, pal): n too large, allowed maximum for palette Set1 is 9
## Returning the palette you asked for with that many colors
TAKEAWAY The effect is similar to what is seen for total arthropods. Spodoptera survival is negatively correlated with herbivore counts which is puzzling for reasons stated above. If a generalist chewing herbivore has a higher chance of survival on a plant, should that not mean more herbivores should be found on the plant in a natural setting??? Definitely a chin-scratcher. As for the interactive effect, most are significant but the estimates vary widely when ACMGLA and STAAJU are excluded from the data set. I need to speak with K$ about this.
CONSIDER Doing another analysis but this time sectioning off chewers from cell suckers since spodoptera are chewers.
2.2.3 log Predator Abundance ~ Spodoptera percent survival
This analysis will determine the strength of the relationship between the percent survival of Spodoptera and predator counts from each plant.
## Run glmmTMB model for Predator abundance ~ Spodoptera percent survival. Family = poisson.
predcount_surv <- glmmTMB(Abundance_Pred ~ Spod_perc_survival * Species + Year + (1|Plot) + (1|Watering_Treatment) + (1|Plant), data = total_plant, family = nbinom2())
##Test dispersion with Dharma package.
testDispersion(predcount_surv)
##
## DHARMa nonparametric dispersion test via sd of residuals fitted vs.
## simulated
##
## data: simulationOutput
## dispersion = 1.116, p-value = 0.44
## alternative hypothesis: two.sided## results indicate the data IS uner-dispersed. Need to switch to negative binomial.
Anova(predcount_surv, type = 3)| Chisq | Df | Pr(>Chisq) |
|---|---|---|
| 41.5 | 1 | 1.2e-10 |
| 0.805 | 1 | 0.37 |
| 209 | 13 | 2.17e-37 |
| 147 | 2 | 1.32e-32 |
| 36.2 | 13 | 0.000544 |
summary(predcount_surv)## Family: nbinom2 ( log )
## Formula:
## Abundance_Pred ~ Spod_perc_survival * Species + Year + (1 | Plot) +
## (1 | Watering_Treatment) + (1 | Plant)
## Data: total_plant
##
## AIC BIC logLik deviance df.resid
## 3377.8 3527.6 -1654.9 3309.8 572
##
## Random effects:
##
## Conditional model:
## Groups Name Variance Std.Dev.
## Plot (Intercept) 0.007098 0.08425
## Watering_Treatment (Intercept) 0.014941 0.12223
## Plant (Intercept) 0.011064 0.10519
## Number of obs: 606, groups: Plot, 20; Watering_Treatment, 6; Plant, 246
##
## Dispersion parameter for nbinom2 family (): 1.87
##
## Conditional model:
## Estimate Std. Error
## (Intercept) 2.31862 0.36007
## Spod_perc_survival -0.55217 0.61560
## SpeciesArtemisia_californica -0.03770 0.37551
## SpeciesDiplacus_aurantiacus -0.19074 0.37716
## SpeciesEncelia_californica -0.48643 0.37320
## SpeciesEriogonum_fasciculatum -0.64813 0.41937
## SpeciesGrindelia_camporum -0.20418 0.42670
## SpeciesIsocoma_menziesii 0.14561 0.38394
## SpeciesMalacothamnus_fasciculatus -0.38332 0.38150
## SpeciesMalacothrix_saxatilis -0.98350 0.40152
## SpeciesMirabilis_laevis -1.97031 0.39234
## SpeciesSalvia_apiana -0.54411 0.39972
## SpeciesSalvia_mellifera -0.43443 0.37988
## SpeciesSisyrinchium_bellum -2.00869 0.41495
## SpeciesStachys_ajugoides_var_rigida -1.71567 0.39796
## Year2022 -0.58501 0.09690
## Year2023 0.62721 0.09066
## Spod_perc_survival:SpeciesArtemisia_californica 0.03542 0.73846
## Spod_perc_survival:SpeciesDiplacus_aurantiacus 0.34321 0.71470
## Spod_perc_survival:SpeciesEncelia_californica -0.40461 1.02699
## Spod_perc_survival:SpeciesEriogonum_fasciculatum 1.06576 0.73024
## Spod_perc_survival:SpeciesGrindelia_camporum 0.34417 0.77934
## Spod_perc_survival:SpeciesIsocoma_menziesii 0.38872 0.75242
## Spod_perc_survival:SpeciesMalacothamnus_fasciculatus 0.91845 0.80631
## Spod_perc_survival:SpeciesMalacothrix_saxatilis -0.09185 0.85585
## Spod_perc_survival:SpeciesMirabilis_laevis 3.45488 1.10263
## Spod_perc_survival:SpeciesSalvia_apiana 0.61749 0.72002
## Spod_perc_survival:SpeciesSalvia_mellifera 0.04658 0.76792
## Spod_perc_survival:SpeciesSisyrinchium_bellum 3.10563 0.87975
## Spod_perc_survival:SpeciesStachys_ajugoides_var_rigida -65.18822 8437.22020
## z value Pr(>|z|)
## (Intercept) 6.439 1.20e-10 ***
## Spod_perc_survival -0.897 0.369740
## SpeciesArtemisia_californica -0.100 0.920021
## SpeciesDiplacus_aurantiacus -0.506 0.613036
## SpeciesEncelia_californica -1.303 0.192434
## SpeciesEriogonum_fasciculatum -1.546 0.122225
## SpeciesGrindelia_camporum -0.479 0.632289
## SpeciesIsocoma_menziesii 0.379 0.704503
## SpeciesMalacothamnus_fasciculatus -1.005 0.315015
## SpeciesMalacothrix_saxatilis -2.449 0.014307 *
## SpeciesMirabilis_laevis -5.022 5.12e-07 ***
## SpeciesSalvia_apiana -1.361 0.173444
## SpeciesSalvia_mellifera -1.144 0.252782
## SpeciesSisyrinchium_bellum -4.841 1.29e-06 ***
## SpeciesStachys_ajugoides_var_rigida -4.311 1.62e-05 ***
## Year2022 -6.037 1.57e-09 ***
## Year2023 6.918 4.57e-12 ***
## Spod_perc_survival:SpeciesArtemisia_californica 0.048 0.961739
## Spod_perc_survival:SpeciesDiplacus_aurantiacus 0.480 0.631077
## Spod_perc_survival:SpeciesEncelia_californica -0.394 0.693603
## Spod_perc_survival:SpeciesEriogonum_fasciculatum 1.459 0.144435
## Spod_perc_survival:SpeciesGrindelia_camporum 0.442 0.658772
## Spod_perc_survival:SpeciesIsocoma_menziesii 0.517 0.605419
## Spod_perc_survival:SpeciesMalacothamnus_fasciculatus 1.139 0.254669
## Spod_perc_survival:SpeciesMalacothrix_saxatilis -0.107 0.914531
## Spod_perc_survival:SpeciesMirabilis_laevis 3.133 0.001728 **
## Spod_perc_survival:SpeciesSalvia_apiana 0.858 0.391110
## Spod_perc_survival:SpeciesSalvia_mellifera 0.061 0.951631
## Spod_perc_survival:SpeciesSisyrinchium_bellum 3.530 0.000415 ***
## Spod_perc_survival:SpeciesStachys_ajugoides_var_rigida -0.008 0.993835
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1## look at plot of model
plot_model(predcount_surv, type = "pred", terms = c("Spod_perc_survival", "Species"), se - T,ci.lvl = .5)+
ggtitle("Predicted total predator counts as a function of spodoptera survival")## Warning in RColorBrewer::brewer.pal(n, pal): n too large, allowed maximum for palette Set1 is 9
## Returning the palette you asked for with that many colors
TAKEAWAY There is little/no relationship between percent survival of spodoptera and total predator counts which, I suppose, is to be expected since spodoptera are not predators but rather herbivores.
Lets look at omnivores….
2.2.4 log Omnivore abundance ~ Spodoptera percent survival
## Run glmmTMB model for Omnivore abundance ~ Spodoptera percent survival. Family = negative binomial.
omnicount_surv <- glmmTMB(Abundance_Omni ~ Spod_perc_survival * Species + Year + (1|Plot) + (1|Watering_Treatment) + (1|Plant), data = total_plant, family = nbinom2())
##Test dispersion with Dharma package.
testDispersion(omnicount_surv)
##
## DHARMa nonparametric dispersion test via sd of residuals fitted vs.
## simulated
##
## data: simulationOutput
## dispersion = 1.2972, p-value = 0.152
## alternative hypothesis: two.sided## results indicate the data IS uner-dispersed. Need to switch to negative binomial. Even with switch, dispersal is not great but 2-sided p-value is >.05
Anova(omnicount_surv, type = 3)| Chisq | Df | Pr(>Chisq) |
|---|---|---|
| 8.3 | 1 | 0.00397 |
| 0.252 | 1 | 0.615 |
| 239 | 13 | 1.11e-43 |
| 75 | 2 | 5.19e-17 |
| 44.9 | 13 | 2.16e-05 |
summary(omnicount_surv)## Family: nbinom2 ( log )
## Formula:
## Abundance_Omni ~ Spod_perc_survival * Species + Year + (1 | Plot) +
## (1 | Watering_Treatment) + (1 | Plant)
## Data: total_plant
##
## AIC BIC logLik deviance df.resid
## 4246.0 4395.8 -2089.0 4178.0 572
##
## Random effects:
##
## Conditional model:
## Groups Name Variance Std.Dev.
## Plot (Intercept) 1.404e-02 1.185e-01
## Watering_Treatment (Intercept) 6.478e-10 2.545e-05
## Plant (Intercept) 1.508e-08 1.228e-04
## Number of obs: 606, groups: Plot, 20; Watering_Treatment, 6; Plant, 246
##
## Dispersion parameter for nbinom2 family (): 1.34
##
## Conditional model:
## Estimate Std. Error
## (Intercept) 1.13505 0.39400
## Spod_perc_survival 0.33897 0.67489
## SpeciesArtemisia_californica 2.35394 0.41850
## SpeciesDiplacus_aurantiacus 1.56960 0.42008
## SpeciesEncelia_californica 1.51054 0.41700
## SpeciesEriogonum_fasciculatum 1.28394 0.45841
## SpeciesGrindelia_camporum 1.37368 0.48375
## SpeciesIsocoma_menziesii 1.04703 0.42881
## SpeciesMalacothamnus_fasciculatus 2.16131 0.42381
## SpeciesMalacothrix_saxatilis 1.02961 0.43956
## SpeciesMirabilis_laevis 1.96932 0.41879
## SpeciesSalvia_apiana 0.78578 0.44092
## SpeciesSalvia_mellifera 0.80383 0.42166
## SpeciesSisyrinchium_bellum -0.13546 0.44797
## SpeciesStachys_ajugoides_var_rigida -0.22850 0.43752
## Year2022 -0.22316 0.10079
## Year2023 0.70253 0.10690
## Spod_perc_survival:SpeciesArtemisia_californica -3.01528 0.81252
## Spod_perc_survival:SpeciesDiplacus_aurantiacus -0.09999 0.78634
## Spod_perc_survival:SpeciesEncelia_californica -1.53250 1.02564
## Spod_perc_survival:SpeciesEriogonum_fasciculatum -0.41629 0.78608
## Spod_perc_survival:SpeciesGrindelia_camporum -2.34563 0.89249
## Spod_perc_survival:SpeciesIsocoma_menziesii -0.39314 0.86615
## Spod_perc_survival:SpeciesMalacothamnus_fasciculatus -0.24848 0.89179
## Spod_perc_survival:SpeciesMalacothrix_saxatilis -1.14143 0.88447
## Spod_perc_survival:SpeciesMirabilis_laevis -0.58757 1.08144
## Spod_perc_survival:SpeciesSalvia_apiana -0.48983 0.78442
## Spod_perc_survival:SpeciesSalvia_mellifera -0.09178 0.83052
## Spod_perc_survival:SpeciesSisyrinchium_bellum 0.82489 1.02046
## Spod_perc_survival:SpeciesStachys_ajugoides_var_rigida 1.80549 2.57608
## z value Pr(>|z|)
## (Intercept) 2.881 0.003966 **
## Spod_perc_survival 0.502 0.615489
## SpeciesArtemisia_californica 5.625 1.86e-08 ***
## SpeciesDiplacus_aurantiacus 3.736 0.000187 ***
## SpeciesEncelia_californica 3.622 0.000292 ***
## SpeciesEriogonum_fasciculatum 2.801 0.005097 **
## SpeciesGrindelia_camporum 2.840 0.004517 **
## SpeciesIsocoma_menziesii 2.442 0.014617 *
## SpeciesMalacothamnus_fasciculatus 5.100 3.40e-07 ***
## SpeciesMalacothrix_saxatilis 2.342 0.019161 *
## SpeciesMirabilis_laevis 4.702 2.57e-06 ***
## SpeciesSalvia_apiana 1.782 0.074725 .
## SpeciesSalvia_mellifera 1.906 0.056607 .
## SpeciesSisyrinchium_bellum -0.302 0.762354
## SpeciesStachys_ajugoides_var_rigida -0.522 0.601484
## Year2022 -2.214 0.026814 *
## Year2023 6.572 4.97e-11 ***
## Spod_perc_survival:SpeciesArtemisia_californica -3.711 0.000206 ***
## Spod_perc_survival:SpeciesDiplacus_aurantiacus -0.127 0.898816
## Spod_perc_survival:SpeciesEncelia_californica -1.494 0.135127
## Spod_perc_survival:SpeciesEriogonum_fasciculatum -0.530 0.596406
## Spod_perc_survival:SpeciesGrindelia_camporum -2.628 0.008584 **
## Spod_perc_survival:SpeciesIsocoma_menziesii -0.454 0.649905
## Spod_perc_survival:SpeciesMalacothamnus_fasciculatus -0.279 0.780525
## Spod_perc_survival:SpeciesMalacothrix_saxatilis -1.291 0.196868
## Spod_perc_survival:SpeciesMirabilis_laevis -0.543 0.586907
## Spod_perc_survival:SpeciesSalvia_apiana -0.624 0.532331
## Spod_perc_survival:SpeciesSalvia_mellifera -0.111 0.912007
## Spod_perc_survival:SpeciesSisyrinchium_bellum 0.808 0.418889
## Spod_perc_survival:SpeciesStachys_ajugoides_var_rigida 0.701 0.483385
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1## look at plot of model
plot_model(omnicount_surv, type = "pred", terms = c("Spod_perc_survival", "Species"), se - T,ci.lvl = .5)+
ggtitle("Predicted total omnivore counts as a function of spodoptera survival")## Warning in RColorBrewer::brewer.pal(n, pal): n too large, allowed maximum for palette Set1 is 9
## Returning the palette you asked for with that many colors
TAKEAWAY There is little/no relationship between percent survival of spodoptera and total omnivore counts, similar to predators when all species are included. However, when I take out ACMGLA and STAAJU, a strong negative relationship overall is present, however with the interactive term, it seems that some species show slight positive while others show slight negative relationships. I don’t quite know what to make of this yet.
Let’s try the detritovores just cause…
2.2.5 log Detritivore abundance ~ Spodoptera percent survival
## Run glmmTMB model for Detritivore abundance ~ Spodoptera percent survival. Family = negative binomial.
detrcount_surv <- glmmTMB(Abundance_Detr ~ Spod_perc_survival * Species + Year + (1|Plot) + (1|Watering_Treatment) + (1|Plant), data = total_plant, family = poisson())
##Test dispersion with Dharma package.
testDispersion(detrcount_surv)
##
## DHARMa nonparametric dispersion test via sd of residuals fitted vs.
## simulated
##
## data: simulationOutput
## dispersion = 0.31661, p-value = 0.752
## alternative hypothesis: two.sided## results indicate the data adequately dispersed and there is no need to switch to negative binomial.
Anova(detrcount_surv, type = 3)| Chisq | Df | Pr(>Chisq) |
|---|---|---|
| 6.88 | 1 | 0.00869 |
| 15.4 | 1 | 8.77e-05 |
| 49.4 | 13 | 3.76e-06 |
| 641 | 2 | 5.48e-140 |
| 196 | 13 | 1.11e-34 |
summary(detrcount_surv)## Family: poisson ( log )
## Formula:
## Abundance_Detr ~ Spod_perc_survival * Species + Year + (1 | Plot) +
## (1 | Watering_Treatment) + (1 | Plant)
## Data: total_plant
##
## AIC BIC logLik deviance df.resid
## 3202.5 3346.1 -1568.2 3136.5 541
##
## Random effects:
##
## Conditional model:
## Groups Name Variance Std.Dev.
## Plot (Intercept) 5.850e-13 7.648e-07
## Watering_Treatment (Intercept) 1.644e-14 1.282e-07
## Plant (Intercept) 1.876e+00 1.370e+00
## Number of obs: 574, groups: Plot, 19; Watering_Treatment, 6; Plant, 233
##
## Conditional model:
## Estimate Std. Error
## (Intercept) -1.41487 0.53923
## Spod_perc_survival 2.31318 0.58975
## SpeciesArtemisia_californica 1.22039 0.64933
## SpeciesDiplacus_aurantiacus 1.54475 0.64294
## SpeciesEncelia_californica 0.96098 0.65984
## SpeciesEriogonum_fasciculatum 1.63229 0.66619
## SpeciesGrindelia_camporum 0.71679 0.79826
## SpeciesIsocoma_menziesii 3.02103 0.65697
## SpeciesMalacothamnus_fasciculatus 1.84861 0.66886
## SpeciesMalacothrix_saxatilis 1.92109 0.69297
## SpeciesMirabilis_laevis 2.14696 0.63513
## SpeciesSalvia_apiana 2.13717 0.66121
## SpeciesSalvia_mellifera 0.47594 0.68699
## SpeciesSisyrinchium_bellum 0.78628 0.70440
## SpeciesStachys_ajugoides_var_rigida 0.42194 0.68926
## Year2022 -2.81733 0.14154
## Year2023 0.56680 0.05979
## Spod_perc_survival:SpeciesArtemisia_californica -5.42835 1.32532
## Spod_perc_survival:SpeciesDiplacus_aurantiacus 0.28939 0.75883
## Spod_perc_survival:SpeciesEncelia_californica -4.31899 2.21411
## Spod_perc_survival:SpeciesEriogonum_fasciculatum -0.31080 0.62788
## Spod_perc_survival:SpeciesGrindelia_camporum -1.27878 1.26083
## Spod_perc_survival:SpeciesIsocoma_menziesii -4.05079 0.66484
## Spod_perc_survival:SpeciesMalacothamnus_fasciculatus 1.31590 1.03852
## Spod_perc_survival:SpeciesMalacothrix_saxatilis -4.20938 1.76347
## Spod_perc_survival:SpeciesMirabilis_laevis 3.88990 0.93304
## Spod_perc_survival:SpeciesSalvia_apiana -1.82259 0.64948
## Spod_perc_survival:SpeciesSalvia_mellifera 0.15829 0.95470
## Spod_perc_survival:SpeciesSisyrinchium_bellum -2.08043 1.43197
## Spod_perc_survival:SpeciesStachys_ajugoides_var_rigida -57.95915 9996.35919
## z value Pr(>|z|)
## (Intercept) -2.624 0.008694 **
## Spod_perc_survival 3.922 8.77e-05 ***
## SpeciesArtemisia_californica 1.879 0.060183 .
## SpeciesDiplacus_aurantiacus 2.403 0.016277 *
## SpeciesEncelia_californica 1.456 0.145286
## SpeciesEriogonum_fasciculatum 2.450 0.014277 *
## SpeciesGrindelia_camporum 0.898 0.369222
## SpeciesIsocoma_menziesii 4.598 4.26e-06 ***
## SpeciesMalacothamnus_fasciculatus 2.764 0.005713 **
## SpeciesMalacothrix_saxatilis 2.772 0.005567 **
## SpeciesMirabilis_laevis 3.380 0.000724 ***
## SpeciesSalvia_apiana 3.232 0.001228 **
## SpeciesSalvia_mellifera 0.693 0.488442
## SpeciesSisyrinchium_bellum 1.116 0.264317
## SpeciesStachys_ajugoides_var_rigida 0.612 0.540431
## Year2022 -19.905 < 2e-16 ***
## Year2023 9.479 < 2e-16 ***
## Spod_perc_survival:SpeciesArtemisia_californica -4.096 4.21e-05 ***
## Spod_perc_survival:SpeciesDiplacus_aurantiacus 0.381 0.702938
## Spod_perc_survival:SpeciesEncelia_californica -1.951 0.051097 .
## Spod_perc_survival:SpeciesEriogonum_fasciculatum -0.495 0.620605
## Spod_perc_survival:SpeciesGrindelia_camporum -1.014 0.310468
## Spod_perc_survival:SpeciesIsocoma_menziesii -6.093 1.11e-09 ***
## Spod_perc_survival:SpeciesMalacothamnus_fasciculatus 1.267 0.205121
## Spod_perc_survival:SpeciesMalacothrix_saxatilis -2.387 0.016987 *
## Spod_perc_survival:SpeciesMirabilis_laevis 4.169 3.06e-05 ***
## Spod_perc_survival:SpeciesSalvia_apiana -2.806 0.005012 **
## Spod_perc_survival:SpeciesSalvia_mellifera 0.166 0.868310
## Spod_perc_survival:SpeciesSisyrinchium_bellum -1.453 0.146267
## Spod_perc_survival:SpeciesStachys_ajugoides_var_rigida -0.006 0.995374
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1## look at plot of model
plot_model(detrcount_surv, type = "pred", terms = c("Spod_perc_survival", "Species"), se - T,ci.lvl = .5)+
ggtitle("Predicted total detritivore counts as a function of spodoptera survival")## Warning in RColorBrewer::brewer.pal(n, pal): n too large, allowed maximum for palette Set1 is 9
## Returning the palette you asked for with that many colors
TAKEAWAY The effect is significant, however in the opposite direction for all species. By itself, the main effect of Spodoptera survival has a positive relationship with the abundance of detritivores. However, the interaction between spodoptera survival and species varies. Some Species show a strong, significant negative effect such as Artemesia. This suggests that on Artemesia, the relationship between spodoptera survival and the abundance of detritivores is not as strong as the overall effect of spodoptera survival on detritivore counts.
Next, I will do the same analyses but looking at level of orders instead of feeding guilds. I am doing this because I want to see if any particular subsets of the feeding guilds seem to be affecting the model more than others. I suspect that this is the case as we observed counts of particulars orders of magnitude greater than other orders, especially those within the same guild, consistently. Furthermore, I suspect ants are having a large effect on the arthropod community given their opportunistic feeding behavior and abundance on certain plants and certain species. Finally, given that the generalist herbivores are chewers, it may turn out that chewers have the expected positive correlation whereas suckers such as aphids and leafhoppers dont due to different feeding behaviors. We shall see…
2.3.1 log Auchennorhyncha ~ Spodoptera survival
## Run glmmTMB model for Auchenorrhyncha abundance ~ Spodoptera percent survival. Family = poisson.
auchcount_surv <- glmmTMB(Abundance_Auchenorrhyncha ~ Spod_perc_survival * Species + Year + (1|Plot) + (1|Watering_Treatment) + (1|Plant), data = total_plant_order, family = poisson())
##Test dispersion with Dharma package.
testDispersion(auchcount_surv)
##
## DHARMa nonparametric dispersion test via sd of residuals fitted vs.
## simulated
##
## data: simulationOutput
## dispersion = 0.53186, p-value = 0.392
## alternative hypothesis: two.sided## results indicate the data is not over- or uner-dispersed and we can stick with poisson and not have to switch to negative binomial.
Anova(auchcount_surv, type = 3)| Chisq | Df | Pr(>Chisq) |
|---|---|---|
| 33 | 1 | 9.36e-09 |
| 4.48 | 1 | 0.0343 |
| 473 | 13 | 9.89e-93 |
| 1.8e+03 | 2 | 0 |
| 317 | 13 | 8.27e-60 |
summary(auchcount_surv)## Family: poisson ( log )
## Formula:
## Abundance_Auchenorrhyncha ~ Spod_perc_survival * Species + Year +
## (1 | Plot) + (1 | Watering_Treatment) + (1 | Plant)
## Data: total_plant_order
##
## AIC BIC logLik deviance df.resid
## 6500.0 6645.4 -3217.0 6434.0 573
##
## Random effects:
##
## Conditional model:
## Groups Name Variance Std.Dev.
## Plot (Intercept) 6.445e-02 2.539e-01
## Watering_Treatment (Intercept) 2.154e-11 4.641e-06
## Plant (Intercept) 6.052e-01 7.779e-01
## Number of obs: 606, groups: Plot, 20; Watering_Treatment, 6; Plant, 246
##
## Conditional model:
## Estimate Std. Error
## (Intercept) 1.86088 0.32408
## Spod_perc_survival -1.06844 0.50487
## SpeciesArtemisia_californica 2.64753 0.36410
## SpeciesDiplacus_aurantiacus -0.58256 0.37878
## SpeciesEncelia_californica 1.18772 0.36648
## SpeciesEriogonum_fasciculatum -0.68333 0.41128
## SpeciesGrindelia_camporum 1.53351 0.38464
## SpeciesIsocoma_menziesii 1.79783 0.37822
## SpeciesMalacothamnus_fasciculatus 0.27022 0.38488
## SpeciesMalacothrix_saxatilis -1.31506 0.42391
## SpeciesMirabilis_laevis 0.75623 0.36953
## SpeciesSalvia_apiana 1.98128 0.37467
## SpeciesSalvia_mellifera 0.83049 0.37249
## SpeciesSisyrinchium_bellum -2.65010 0.48002
## SpeciesStachys_ajugoides_var_rigida -0.75484 0.38728
## Year2022 -1.29908 0.03198
## Year2023 -0.54912 0.02666
## Spod_perc_survival:SpeciesArtemisia_californica 0.83425 0.51276
## Spod_perc_survival:SpeciesDiplacus_aurantiacus 1.89626 0.60086
## Spod_perc_survival:SpeciesEncelia_californica -2.05611 0.73509
## Spod_perc_survival:SpeciesEriogonum_fasciculatum 1.99557 0.57619
## Spod_perc_survival:SpeciesGrindelia_camporum -0.69262 0.59994
## Spod_perc_survival:SpeciesIsocoma_menziesii -1.41880 0.58151
## Spod_perc_survival:SpeciesMalacothamnus_fasciculatus 0.69316 0.68846
## Spod_perc_survival:SpeciesMalacothrix_saxatilis 1.75662 0.81067
## Spod_perc_survival:SpeciesMirabilis_laevis 5.44065 0.65638
## Spod_perc_survival:SpeciesSalvia_apiana -0.16204 0.52041
## Spod_perc_survival:SpeciesSalvia_mellifera 1.26628 0.54482
## Spod_perc_survival:SpeciesSisyrinchium_bellum 2.23572 1.12112
## Spod_perc_survival:SpeciesStachys_ajugoides_var_rigida -65.29126 6116.52208
## z value Pr(>|z|)
## (Intercept) 5.74 9.36e-09 ***
## Spod_perc_survival -2.12 0.034322 *
## SpeciesArtemisia_californica 7.27 3.56e-13 ***
## SpeciesDiplacus_aurantiacus -1.54 0.124054
## SpeciesEncelia_californica 3.24 0.001192 **
## SpeciesEriogonum_fasciculatum -1.66 0.096621 .
## SpeciesGrindelia_camporum 3.99 6.70e-05 ***
## SpeciesIsocoma_menziesii 4.75 2.00e-06 ***
## SpeciesMalacothamnus_fasciculatus 0.70 0.482622
## SpeciesMalacothrix_saxatilis -3.10 0.001921 **
## SpeciesMirabilis_laevis 2.05 0.040709 *
## SpeciesSalvia_apiana 5.29 1.24e-07 ***
## SpeciesSalvia_mellifera 2.23 0.025775 *
## SpeciesSisyrinchium_bellum -5.52 3.38e-08 ***
## SpeciesStachys_ajugoides_var_rigida -1.95 0.051283 .
## Year2022 -40.62 < 2e-16 ***
## Year2023 -20.60 < 2e-16 ***
## Spod_perc_survival:SpeciesArtemisia_californica 1.63 0.103742
## Spod_perc_survival:SpeciesDiplacus_aurantiacus 3.16 0.001600 **
## Spod_perc_survival:SpeciesEncelia_californica -2.80 0.005156 **
## Spod_perc_survival:SpeciesEriogonum_fasciculatum 3.46 0.000533 ***
## Spod_perc_survival:SpeciesGrindelia_camporum -1.15 0.248301
## Spod_perc_survival:SpeciesIsocoma_menziesii -2.44 0.014693 *
## Spod_perc_survival:SpeciesMalacothamnus_fasciculatus 1.01 0.314024
## Spod_perc_survival:SpeciesMalacothrix_saxatilis 2.17 0.030244 *
## Spod_perc_survival:SpeciesMirabilis_laevis 8.29 < 2e-16 ***
## Spod_perc_survival:SpeciesSalvia_apiana -0.31 0.755520
## Spod_perc_survival:SpeciesSalvia_mellifera 2.32 0.020114 *
## Spod_perc_survival:SpeciesSisyrinchium_bellum 1.99 0.046131 *
## Spod_perc_survival:SpeciesStachys_ajugoides_var_rigida -0.01 0.991483
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1## look at plot of model
plot_model(auchcount_surv, type = "pred", terms = c("Spod_perc_survival", "Species"), se - T,ci.lvl = .5)+
ggtitle("Predicted total auchenorrhyncha counts as a function of spodoptera survival")## Warning in RColorBrewer::brewer.pal(n, pal): n too large, allowed maximum for palette Set1 is 9
## Returning the palette you asked for with that many colors
TAKEAWAY A somewhat significant relationship exists between spodoptera survival and auchenorrhyncha counts, however the interactive effect is much more pronounced. Looking at the graph, it seems that the relationship is negative such that increased spodoptera survival correlates with decreased abundance.
2.3.2 log Sternorrhyncha ~ Spodoptera survival
## Run glmmTMB model for Sternorrhyncha abundance ~ Spodoptera percent survival. Family = poisson.
sterncount_surv <- glmmTMB(Abundance_Sternorrhyncha ~ Spod_perc_survival * Species + Year + (1|Plot) + (1|Watering_Treatment) + (1|Plant), data = total_plant_order, family = poisson())## Warning in finalizeTMB(TMBStruc, obj, fit, h, data.tmb.old): Model convergence
## problem; singular convergence (7). See vignette('troubleshooting'),
## help('diagnose')##Test dispersion with Dharma package.
testDispersion(sterncount_surv)
##
## DHARMa nonparametric dispersion test via sd of residuals fitted vs.
## simulated
##
## data: simulationOutput
## dispersion = 0.21797, p-value = 0.656
## alternative hypothesis: two.sided## results indicate the data is not over- or uner-dispersed and we can stick with poisson and not have to switch to negative binomial.
Anova(sterncount_surv, type = 3)| Chisq | Df | Pr(>Chisq) |
|---|---|---|
| 249 | 1 | 3.34e-56 |
| 1.15e+03 | 1 | 8.62e-253 |
| 210 | 13 | 1.14e-37 |
| 3.04e+03 | 2 | 0 |
| 887 | 13 | 3.07e-181 |
summary(sterncount_surv)## Family: poisson ( log )
## Formula:
## Abundance_Sternorrhyncha ~ Spod_perc_survival * Species + Year +
## (1 | Plot) + (1 | Watering_Treatment) + (1 | Plant)
## Data: total_plant_order
##
## AIC BIC logLik deviance df.resid
## 10300.5 10446.0 -5117.3 10234.5 573
##
## Random effects:
##
## Conditional model:
## Groups Name Variance Std.Dev.
## Plot (Intercept) 3.839e-10 1.959e-05
## Watering_Treatment (Intercept) 5.568e-15 7.462e-08
## Plant (Intercept) 2.103e+00 1.450e+00
## Number of obs: 606, groups: Plot, 20; Watering_Treatment, 6; Plant, 246
##
## Conditional model:
## Estimate Std. Error
## (Intercept) 6.06500 0.38397
## Spod_perc_survival -4.98375 0.14675
## SpeciesArtemisia_californica -4.96148 0.52232
## SpeciesDiplacus_aurantiacus -4.79720 0.51671
## SpeciesEncelia_californica -3.73746 0.50925
## SpeciesEriogonum_fasciculatum -4.99456 0.55867
## SpeciesGrindelia_camporum -3.63856 0.53893
## SpeciesIsocoma_menziesii -3.25449 0.53673
## SpeciesMalacothamnus_fasciculatus -3.95749 0.53793
## SpeciesMalacothrix_saxatilis -5.96863 0.59736
## SpeciesMirabilis_laevis -5.26571 0.52864
## SpeciesSalvia_apiana -2.90632 0.52638
## SpeciesSalvia_mellifera -2.24593 0.51706
## SpeciesSisyrinchium_bellum -5.42085 0.56469
## SpeciesStachys_ajugoides_var_rigida -4.43894 0.52195
## Year2022 -2.33848 0.04499
## Year2023 -0.83325 0.03039
## Spod_perc_survival:SpeciesArtemisia_californica 6.01904 0.47219
## Spod_perc_survival:SpeciesDiplacus_aurantiacus 6.21098 0.32136
## Spod_perc_survival:SpeciesEncelia_californica 2.15856 0.73891
## Spod_perc_survival:SpeciesEriogonum_fasciculatum 6.52229 0.32948
## Spod_perc_survival:SpeciesGrindelia_camporum 4.75053 0.30728
## Spod_perc_survival:SpeciesIsocoma_menziesii 4.57226 0.33835
## Spod_perc_survival:SpeciesMalacothamnus_fasciculatus 4.91264 0.55480
## Spod_perc_survival:SpeciesMalacothrix_saxatilis 4.52425 1.08003
## Spod_perc_survival:SpeciesMirabilis_laevis 2.64111 0.75229
## Spod_perc_survival:SpeciesSalvia_apiana 4.12456 0.19419
## Spod_perc_survival:SpeciesSalvia_mellifera 2.07861 0.27889
## Spod_perc_survival:SpeciesSisyrinchium_bellum 4.55748 0.56259
## Spod_perc_survival:SpeciesStachys_ajugoides_var_rigida -53.38258 4373.12660
## z value Pr(>|z|)
## (Intercept) 15.80 < 2e-16 ***
## Spod_perc_survival -33.96 < 2e-16 ***
## SpeciesArtemisia_californica -9.50 < 2e-16 ***
## SpeciesDiplacus_aurantiacus -9.28 < 2e-16 ***
## SpeciesEncelia_californica -7.34 2.15e-13 ***
## SpeciesEriogonum_fasciculatum -8.94 < 2e-16 ***
## SpeciesGrindelia_camporum -6.75 1.46e-11 ***
## SpeciesIsocoma_menziesii -6.06 1.33e-09 ***
## SpeciesMalacothamnus_fasciculatus -7.36 1.88e-13 ***
## SpeciesMalacothrix_saxatilis -9.99 < 2e-16 ***
## SpeciesMirabilis_laevis -9.96 < 2e-16 ***
## SpeciesSalvia_apiana -5.52 3.37e-08 ***
## SpeciesSalvia_mellifera -4.34 1.40e-05 ***
## SpeciesSisyrinchium_bellum -9.60 < 2e-16 ***
## SpeciesStachys_ajugoides_var_rigida -8.50 < 2e-16 ***
## Year2022 -51.98 < 2e-16 ***
## Year2023 -27.42 < 2e-16 ***
## Spod_perc_survival:SpeciesArtemisia_californica 12.75 < 2e-16 ***
## Spod_perc_survival:SpeciesDiplacus_aurantiacus 19.33 < 2e-16 ***
## Spod_perc_survival:SpeciesEncelia_californica 2.92 0.003486 **
## Spod_perc_survival:SpeciesEriogonum_fasciculatum 19.80 < 2e-16 ***
## Spod_perc_survival:SpeciesGrindelia_camporum 15.46 < 2e-16 ***
## Spod_perc_survival:SpeciesIsocoma_menziesii 13.51 < 2e-16 ***
## Spod_perc_survival:SpeciesMalacothamnus_fasciculatus 8.85 < 2e-16 ***
## Spod_perc_survival:SpeciesMalacothrix_saxatilis 4.19 2.80e-05 ***
## Spod_perc_survival:SpeciesMirabilis_laevis 3.51 0.000447 ***
## Spod_perc_survival:SpeciesSalvia_apiana 21.24 < 2e-16 ***
## Spod_perc_survival:SpeciesSalvia_mellifera 7.45 9.11e-14 ***
## Spod_perc_survival:SpeciesSisyrinchium_bellum 8.10 5.46e-16 ***
## Spod_perc_survival:SpeciesStachys_ajugoides_var_rigida -0.01 0.990260
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1## look at plot of model
plot_model(sterncount_surv, type = "pred", terms = c("Spod_perc_survival", "Species"), se - T,ci.lvl = .5)+
ggtitle("Predicted total sternorrhyncha counts as a function of spodoptera survival")## Warning in RColorBrewer::brewer.pal(n, pal): n too large, allowed maximum for palette Set1 is 9
## Returning the palette you asked for with that many colors
TAKEAWAY Unlike auchenorrhyncha, sternorrhyncha show a strong significant relationship with spodoptera survival, and this relationship is negative. The interactive effects are also strongly significant.
2.3.3 log Caterpillars ~ Spodoptera survival
## Run glmmTMB model for Caterpillar abundance ~ Spodoptera percent survival. Family = poisson.
catcount_surv <- glmmTMB(Abundance_Lepidoptera_juvenile ~ Spod_perc_survival * Species + Year + (1|Plot) + (1|Watering_Treatment) + (1|Plant), data = total_plant_order, family = poisson())
##Test dispersion with Dharma package.
testDispersion(catcount_surv)
##
## DHARMa nonparametric dispersion test via sd of residuals fitted vs.
## simulated
##
## data: simulationOutput
## dispersion = 1.3015, p-value = 0.328
## alternative hypothesis: two.sided## results indicate the data is not over- or uner-dispersed and we can stick with poisson and not have to switch to negative binomial.
Anova(catcount_surv, type = 3)| Chisq | Df | Pr(>Chisq) |
|---|---|---|
| 9.1 | 1 | 0.00256 |
| 0.0339 | 1 | 0.854 |
| 64.1 | 13 | 9.72e-09 |
| 236 | 2 | 4.89e-52 |
| 17.4 | 13 | 0.183 |
summary(catcount_surv)## Family: poisson ( log )
## Formula:
## Abundance_Lepidoptera_juvenile ~ Spod_perc_survival * Species +
## Year + (1 | Plot) + (1 | Watering_Treatment) + (1 | Plant)
## Data: total_plant_order
##
## AIC BIC logLik deviance df.resid
## 855.0 1000.5 -394.5 789.0 573
##
## Random effects:
##
## Conditional model:
## Groups Name Variance Std.Dev.
## Plot (Intercept) 7.102e-31 8.427e-16
## Watering_Treatment (Intercept) 2.286e-02 1.512e-01
## Plant (Intercept) 5.860e-01 7.655e-01
## Number of obs: 606, groups: Plot, 20; Watering_Treatment, 6; Plant, 246
##
## Conditional model:
## Estimate Std. Error
## (Intercept) -1.925e+00 6.383e-01
## Spod_perc_survival 2.307e-01 1.252e+00
## SpeciesArtemisia_californica -1.520e+00 7.233e-01
## SpeciesDiplacus_aurantiacus -2.681e+00 8.820e-01
## SpeciesEncelia_californica 6.036e-01 6.690e-01
## SpeciesEriogonum_fasciculatum -9.214e-01 8.898e-01
## SpeciesGrindelia_camporum 5.631e-01 8.844e-01
## SpeciesIsocoma_menziesii -5.862e-01 8.036e-01
## SpeciesMalacothamnus_fasciculatus -1.987e-01 7.267e-01
## SpeciesMalacothrix_saxatilis -1.020e+00 7.918e-01
## SpeciesMirabilis_laevis 4.360e-01 6.698e-01
## SpeciesSalvia_apiana -1.168e+00 9.901e-01
## SpeciesSalvia_mellifera 2.250e-02 6.914e-01
## SpeciesSisyrinchium_bellum -2.150e+00 1.011e+00
## SpeciesStachys_ajugoides_var_rigida -5.423e-01 7.318e-01
## Year2022 -8.392e-01 3.631e-01
## Year2023 2.579e+00 2.054e-01
## Spod_perc_survival:SpeciesArtemisia_californica 1.422e+00 1.996e+00
## Spod_perc_survival:SpeciesDiplacus_aurantiacus 2.312e+00 1.672e+00
## Spod_perc_survival:SpeciesEncelia_californica -9.162e-01 2.126e+00
## Spod_perc_survival:SpeciesEriogonum_fasciculatum -6.801e-01 1.623e+00
## Spod_perc_survival:SpeciesGrindelia_camporum -4.777e+00 3.328e+00
## Spod_perc_survival:SpeciesIsocoma_menziesii -2.251e+00 1.954e+00
## Spod_perc_survival:SpeciesMalacothamnus_fasciculatus -9.196e-01 2.007e+00
## Spod_perc_survival:SpeciesMalacothrix_saxatilis 4.651e-02 1.905e+00
## Spod_perc_survival:SpeciesMirabilis_laevis -2.444e+00 2.507e+00
## Spod_perc_survival:SpeciesSalvia_apiana -2.704e+00 2.191e+00
## Spod_perc_survival:SpeciesSalvia_mellifera 7.337e-01 1.429e+00
## Spod_perc_survival:SpeciesSisyrinchium_bellum 2.403e+00 2.112e+00
## Spod_perc_survival:SpeciesStachys_ajugoides_var_rigida -5.995e+01 2.179e+04
## z value Pr(>|z|)
## (Intercept) -3.016 0.00256 **
## Spod_perc_survival 0.184 0.85387
## SpeciesArtemisia_californica -2.101 0.03562 *
## SpeciesDiplacus_aurantiacus -3.039 0.00237 **
## SpeciesEncelia_californica 0.902 0.36694
## SpeciesEriogonum_fasciculatum -1.036 0.30042
## SpeciesGrindelia_camporum 0.637 0.52430
## SpeciesIsocoma_menziesii -0.729 0.46574
## SpeciesMalacothamnus_fasciculatus -0.273 0.78449
## SpeciesMalacothrix_saxatilis -1.288 0.19782
## SpeciesMirabilis_laevis 0.651 0.51505
## SpeciesSalvia_apiana -1.180 0.23810
## SpeciesSalvia_mellifera 0.033 0.97404
## SpeciesSisyrinchium_bellum -2.126 0.03350 *
## SpeciesStachys_ajugoides_var_rigida -0.741 0.45867
## Year2022 -2.311 0.02083 *
## Year2023 12.557 < 2e-16 ***
## Spod_perc_survival:SpeciesArtemisia_californica 0.713 0.47609
## Spod_perc_survival:SpeciesDiplacus_aurantiacus 1.383 0.16660
## Spod_perc_survival:SpeciesEncelia_californica -0.431 0.66650
## Spod_perc_survival:SpeciesEriogonum_fasciculatum -0.419 0.67511
## Spod_perc_survival:SpeciesGrindelia_camporum -1.436 0.15111
## Spod_perc_survival:SpeciesIsocoma_menziesii -1.152 0.24938
## Spod_perc_survival:SpeciesMalacothamnus_fasciculatus -0.458 0.64678
## Spod_perc_survival:SpeciesMalacothrix_saxatilis 0.024 0.98052
## Spod_perc_survival:SpeciesMirabilis_laevis -0.975 0.32976
## Spod_perc_survival:SpeciesSalvia_apiana -1.234 0.21712
## Spod_perc_survival:SpeciesSalvia_mellifera 0.514 0.60756
## Spod_perc_survival:SpeciesSisyrinchium_bellum 1.138 0.25520
## Spod_perc_survival:SpeciesStachys_ajugoides_var_rigida -0.003 0.99780
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1## look at plot of model
plot_model(catcount_surv, type = "pred", terms = c("Spod_perc_survival", "Species"), se - T,ci.lvl = .5)+
ggtitle("Predicted total caterpillar counts as a function of spodoptera survival")## Warning in RColorBrewer::brewer.pal(n, pal): n too large, allowed maximum for palette Set1 is 9
## Returning the palette you asked for with that many colors
TAKEAWAY Caterpillars do not show any significant relationship with spodoptera survival, somewhat surprisingly. The model is a bit all over the place when predicting total counts with some slopes positive and others negative.
2.3.4 log Heteroptera ~ Spodoptera survival
## Run glmmTMB model for Heteroptera abundance ~ Spodoptera percent survival. Family = poisson.
hetcount_surv <- glmmTMB(Abundance_Heteroptera ~ Spod_perc_survival * Species + Year + (1|Plot) + (1|Watering_Treatment) + (1|Plant), data = total_plant_order, family = poisson())
##Test dispersion with Dharma package.
testDispersion(hetcount_surv)
##
## DHARMa nonparametric dispersion test via sd of residuals fitted vs.
## simulated
##
## data: simulationOutput
## dispersion = 0.84105, p-value = 0.824
## alternative hypothesis: two.sided## results indicate the data is not over- or uner-dispersed and we can stick with poisson and not have to switch to negative binomial.
Anova(hetcount_surv, type = 3)| Chisq | Df | Pr(>Chisq) |
|---|---|---|
| 2.76 | 1 | 0.0966 |
| 0.0981 | 1 | 0.754 |
| 372 | 13 | 1.81e-71 |
| 699 | 2 | 1.28e-152 |
| 277 | 13 | 1.76e-51 |
summary(hetcount_surv)## Family: poisson ( log )
## Formula:
## Abundance_Heteroptera ~ Spod_perc_survival * Species + Year +
## (1 | Plot) + (1 | Watering_Treatment) + (1 | Plant)
## Data: total_plant_order
##
## AIC BIC logLik deviance df.resid
## 5842.7 5988.0 -2888.3 5776.7 572
##
## Random effects:
##
## Conditional model:
## Groups Name Variance Std.Dev.
## Plot (Intercept) 1.928e-08 0.0001388
## Watering_Treatment (Intercept) 9.670e-10 0.0000311
## Plant (Intercept) 6.187e-01 0.7865662
## Number of obs: 605, groups: Plot, 20; Watering_Treatment, 6; Plant, 246
##
## Conditional model:
## Estimate Std. Error
## (Intercept) -0.84092 0.50612
## Spod_perc_survival 0.24530 0.78305
## SpeciesArtemisia_californica 4.23552 0.53637
## SpeciesDiplacus_aurantiacus 1.89513 0.54257
## SpeciesEncelia_californica 0.56108 0.55636
## SpeciesEriogonum_fasciculatum 1.17223 0.57667
## SpeciesGrindelia_camporum 2.22589 0.58028
## SpeciesIsocoma_menziesii 2.08987 0.55274
## SpeciesMalacothamnus_fasciculatus 3.05489 0.54640
## SpeciesMalacothrix_saxatilis 0.58193 0.58287
## SpeciesMirabilis_laevis 2.73267 0.53849
## SpeciesSalvia_apiana 1.87205 0.55952
## SpeciesSalvia_mellifera 1.38777 0.55088
## SpeciesSisyrinchium_bellum -0.66873 0.65079
## SpeciesStachys_ajugoides_var_rigida -0.03705 0.58690
## Year2022 0.06331 0.04791
## Year2023 0.89222 0.03950
## Spod_perc_survival:SpeciesArtemisia_californica -4.04266 0.80899
## Spod_perc_survival:SpeciesDiplacus_aurantiacus -0.09705 0.82533
## Spod_perc_survival:SpeciesEncelia_californica -3.72520 1.63629
## Spod_perc_survival:SpeciesEriogonum_fasciculatum -0.05387 0.85051
## Spod_perc_survival:SpeciesGrindelia_camporum -2.74261 1.01978
## Spod_perc_survival:SpeciesIsocoma_menziesii -0.91384 0.84183
## Spod_perc_survival:SpeciesMalacothamnus_fasciculatus 0.08430 0.83856
## Spod_perc_survival:SpeciesMalacothrix_saxatilis 0.12765 0.99866
## Spod_perc_survival:SpeciesMirabilis_laevis -0.99809 0.90078
## Spod_perc_survival:SpeciesSalvia_apiana -2.93914 0.88954
## Spod_perc_survival:SpeciesSalvia_mellifera -0.96627 0.88824
## Spod_perc_survival:SpeciesSisyrinchium_bellum -0.52555 1.45929
## Spod_perc_survival:SpeciesStachys_ajugoides_var_rigida 7.15472 2.06656
## z value Pr(>|z|)
## (Intercept) -1.661 0.096616 .
## Spod_perc_survival 0.313 0.754078
## SpeciesArtemisia_californica 7.897 2.87e-15 ***
## SpeciesDiplacus_aurantiacus 3.493 0.000478 ***
## SpeciesEncelia_californica 1.008 0.313218
## SpeciesEriogonum_fasciculatum 2.033 0.042077 *
## SpeciesGrindelia_camporum 3.836 0.000125 ***
## SpeciesIsocoma_menziesii 3.781 0.000156 ***
## SpeciesMalacothamnus_fasciculatus 5.591 2.26e-08 ***
## SpeciesMalacothrix_saxatilis 0.998 0.318093
## SpeciesMirabilis_laevis 5.075 3.88e-07 ***
## SpeciesSalvia_apiana 3.346 0.000820 ***
## SpeciesSalvia_mellifera 2.519 0.011762 *
## SpeciesSisyrinchium_bellum -1.028 0.304157
## SpeciesStachys_ajugoides_var_rigida -0.063 0.949664
## Year2022 1.321 0.186352
## Year2023 22.589 < 2e-16 ***
## Spod_perc_survival:SpeciesArtemisia_californica -4.997 5.82e-07 ***
## Spod_perc_survival:SpeciesDiplacus_aurantiacus -0.118 0.906394
## Spod_perc_survival:SpeciesEncelia_californica -2.277 0.022810 *
## Spod_perc_survival:SpeciesEriogonum_fasciculatum -0.063 0.949493
## Spod_perc_survival:SpeciesGrindelia_camporum -2.689 0.007158 **
## Spod_perc_survival:SpeciesIsocoma_menziesii -1.086 0.277682
## Spod_perc_survival:SpeciesMalacothamnus_fasciculatus 0.101 0.919921
## Spod_perc_survival:SpeciesMalacothrix_saxatilis 0.128 0.898292
## Spod_perc_survival:SpeciesMirabilis_laevis -1.108 0.267846
## Spod_perc_survival:SpeciesSalvia_apiana -3.304 0.000953 ***
## Spod_perc_survival:SpeciesSalvia_mellifera -1.088 0.276663
## Spod_perc_survival:SpeciesSisyrinchium_bellum -0.360 0.718741
## Spod_perc_survival:SpeciesStachys_ajugoides_var_rigida 3.462 0.000536 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1## look at plot of model
plot_model(hetcount_surv, type = "pred", terms = c("Spod_perc_survival", "Species"), se - T,ci.lvl = .5)+
ggtitle("Predicted total heteroptera counts as a function of spodoptera survival")## Warning in RColorBrewer::brewer.pal(n, pal): n too large, allowed maximum for palette Set1 is 9
## Returning the palette you asked for with that many colors
TAKEAWAY No significant effect alone, but an interactive effect of percent survival * species. One species is obscuring the scale of the graph so I will need to eliminate that species in order to get a better idea on whats happening.
2.3.5 log Ant count ~ Spodoptera survival
antcount_surv <- glmmTMB(Abundance_Hymenoptera_Formicidae ~ Spod_perc_survival * Species + Year + (1|Plot) + (1|Watering_Treatment) + (1|Plant), data = total_plant_order, family = poisson())
##Test dispersion with Dharma package.
testDispersion(antcount_surv)
##
## DHARMa nonparametric dispersion test via sd of residuals fitted vs.
## simulated
##
## data: simulationOutput
## dispersion = 0.49657, p-value = 0.848
## alternative hypothesis: two.sided## results indicate the data is not over- or uner-dispersed and we can stick with poisson and not have to switch to negative binomial.
Anova(antcount_surv, type = 3)| Chisq | Df | Pr(>Chisq) |
|---|---|---|
| 26.2 | 1 | 3.1e-07 |
| 80.5 | 1 | 2.95e-19 |
| 181 | 13 | 8.83e-32 |
| 1.36e+03 | 2 | 1.33e-295 |
| 493 | 13 | 4.04e-97 |
summary(antcount_surv)## Family: poisson ( log )
## Formula:
## Abundance_Hymenoptera_Formicidae ~ Spod_perc_survival * Species +
## Year + (1 | Plot) + (1 | Watering_Treatment) + (1 | Plant)
## Data: total_plant_order
##
## AIC BIC logLik deviance df.resid
## 8253.7 8399.1 -4093.9 8187.7 573
##
## Random effects:
##
## Conditional model:
## Groups Name Variance Std.Dev.
## Plot (Intercept) 1.245e-02 0.1115875
## Watering_Treatment (Intercept) 2.809e-08 0.0001676
## Plant (Intercept) 1.401e+00 1.1835366
## Number of obs: 606, groups: Plot, 20; Watering_Treatment, 6; Plant, 246
##
## Conditional model:
## Estimate Std. Error
## (Intercept) 2.39169 0.46739
## Spod_perc_survival -7.39596 0.82448
## SpeciesArtemisia_californica -1.20169 0.55031
## SpeciesDiplacus_aurantiacus -1.12978 0.54801
## SpeciesEncelia_californica -0.41380 0.54351
## SpeciesEriogonum_fasciculatum -0.48250 0.56252
## SpeciesGrindelia_camporum -2.38254 0.60840
## SpeciesIsocoma_menziesii -0.99704 0.56678
## SpeciesMalacothamnus_fasciculatus 0.28457 0.55642
## SpeciesMalacothrix_saxatilis -2.01909 0.59806
## SpeciesMirabilis_laevis -1.88836 0.55834
## SpeciesSalvia_apiana 1.48730 0.55055
## SpeciesSalvia_mellifera -0.24590 0.54780
## SpeciesSisyrinchium_bellum -2.11597 0.58298
## SpeciesStachys_ajugoides_var_rigida -3.51583 0.62486
## Year2022 0.30908 0.02816
## Year2023 -1.08867 0.03883
## Spod_perc_survival:SpeciesArtemisia_californica 7.45910 0.90222
## Spod_perc_survival:SpeciesDiplacus_aurantiacus 8.70110 0.85445
## Spod_perc_survival:SpeciesEncelia_californica 1.81547 1.00962
## Spod_perc_survival:SpeciesEriogonum_fasciculatum 8.22797 0.84117
## Spod_perc_survival:SpeciesGrindelia_camporum 7.21526 0.92974
## Spod_perc_survival:SpeciesIsocoma_menziesii 6.05610 0.91998
## Spod_perc_survival:SpeciesMalacothamnus_fasciculatus 9.33759 0.84653
## Spod_perc_survival:SpeciesMalacothrix_saxatilis 6.66802 0.85633
## Spod_perc_survival:SpeciesMirabilis_laevis 1.21021 1.30253
## Spod_perc_survival:SpeciesSalvia_apiana 6.77152 0.82632
## Spod_perc_survival:SpeciesSalvia_mellifera 8.29102 0.84536
## Spod_perc_survival:SpeciesSisyrinchium_bellum 5.39389 1.07890
## Spod_perc_survival:SpeciesStachys_ajugoides_var_rigida 11.71856 2.59233
## z value Pr(>|z|)
## (Intercept) 5.117 3.10e-07 ***
## Spod_perc_survival -8.970 < 2e-16 ***
## SpeciesArtemisia_californica -2.184 0.028987 *
## SpeciesDiplacus_aurantiacus -2.062 0.039244 *
## SpeciesEncelia_californica -0.761 0.446448
## SpeciesEriogonum_fasciculatum -0.858 0.391037
## SpeciesGrindelia_camporum -3.916 9.00e-05 ***
## SpeciesIsocoma_menziesii -1.759 0.078556 .
## SpeciesMalacothamnus_fasciculatus 0.511 0.609046
## SpeciesMalacothrix_saxatilis -3.376 0.000735 ***
## SpeciesMirabilis_laevis -3.382 0.000719 ***
## SpeciesSalvia_apiana 2.701 0.006903 **
## SpeciesSalvia_mellifera -0.449 0.653510
## SpeciesSisyrinchium_bellum -3.630 0.000284 ***
## SpeciesStachys_ajugoides_var_rigida -5.627 1.84e-08 ***
## Year2022 10.978 < 2e-16 ***
## Year2023 -28.038 < 2e-16 ***
## Spod_perc_survival:SpeciesArtemisia_californica 8.268 < 2e-16 ***
## Spod_perc_survival:SpeciesDiplacus_aurantiacus 10.183 < 2e-16 ***
## Spod_perc_survival:SpeciesEncelia_californica 1.798 0.072149 .
## Spod_perc_survival:SpeciesEriogonum_fasciculatum 9.782 < 2e-16 ***
## Spod_perc_survival:SpeciesGrindelia_camporum 7.761 8.46e-15 ***
## Spod_perc_survival:SpeciesIsocoma_menziesii 6.583 4.62e-11 ***
## Spod_perc_survival:SpeciesMalacothamnus_fasciculatus 11.030 < 2e-16 ***
## Spod_perc_survival:SpeciesMalacothrix_saxatilis 7.787 6.87e-15 ***
## Spod_perc_survival:SpeciesMirabilis_laevis 0.929 0.352823
## Spod_perc_survival:SpeciesSalvia_apiana 8.195 2.51e-16 ***
## Spod_perc_survival:SpeciesSalvia_mellifera 9.808 < 2e-16 ***
## Spod_perc_survival:SpeciesSisyrinchium_bellum 4.999 5.75e-07 ***
## Spod_perc_survival:SpeciesStachys_ajugoides_var_rigida 4.520 6.17e-06 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1## look at plot of model
plot_model(antcount_surv, type = "pred", terms = c("Spod_perc_survival", "Species"), se - T,ci.lvl = .5)+
ggtitle("Predicted total ant counts as a function of spodoptera survival")## Warning in RColorBrewer::brewer.pal(n, pal): n too large, allowed maximum for palette Set1 is 9
## Returning the palette you asked for with that many colors
TAKEAWAY Ants do show a significant relationship with spodoptera survival 2.3.6 log Wasp count ~ Spodoptera survival
waspcount_surv <- glmmTMB(Abundance_Hymenoptera_Vespidae ~ Spod_perc_survival * Species + Year + (1|Plot) + (1|Watering_Treatment) + (1|Plant), data = total_plant_order, family = nbinom2())
##Test dispersion with Dharma package.
testDispersion(waspcount_surv)
##
## DHARMa nonparametric dispersion test via sd of residuals fitted vs.
## simulated
##
## data: simulationOutput
## dispersion = 1.0495, p-value = 0.68
## alternative hypothesis: two.sided## results indicate the data is over-dispersed and we need to switch to negative binomial.
Anova(waspcount_surv, type = 3)| Chisq | Df | Pr(>Chisq) |
|---|---|---|
| 26 | 1 | 3.43e-07 |
| 0.00158 | 1 | 0.968 |
| 135 | 13 | 2.34e-22 |
| 110 | 2 | 1.22e-24 |
| 24.2 | 13 | 0.0297 |
summary(waspcount_surv)## Family: nbinom2 ( log )
## Formula:
## Abundance_Hymenoptera_Vespidae ~ Spod_perc_survival * Species +
## Year + (1 | Plot) + (1 | Watering_Treatment) + (1 | Plant)
## Data: total_plant_order
##
## AIC BIC logLik deviance df.resid
## 2547.5 2697.3 -1239.8 2479.5 571
##
## Random effects:
##
## Conditional model:
## Groups Name Variance Std.Dev.
## Plot (Intercept) 9.843e-20 3.137e-10
## Watering_Treatment (Intercept) 2.892e-03 5.378e-02
## Plant (Intercept) 6.720e-18 2.592e-09
## Number of obs: 605, groups: Plot, 20; Watering_Treatment, 6; Plant, 246
##
## Dispersion parameter for nbinom2 family (): 1.53
##
## Conditional model:
## Estimate Std. Error
## (Intercept) 1.96801 0.38602
## Spod_perc_survival 0.02755 0.69217
## SpeciesArtemisia_californica -0.48493 0.41548
## SpeciesDiplacus_aurantiacus -0.17779 0.41377
## SpeciesEncelia_californica -0.38027 0.41199
## SpeciesEriogonum_fasciculatum -0.67074 0.47872
## SpeciesGrindelia_camporum -0.60255 0.48897
## SpeciesIsocoma_menziesii -0.49462 0.42808
## SpeciesMalacothamnus_fasciculatus -0.84849 0.42734
## SpeciesMalacothrix_saxatilis -1.16724 0.44817
## SpeciesMirabilis_laevis -1.93506 0.43697
## SpeciesSalvia_apiana -0.59085 0.44162
## SpeciesSalvia_mellifera -0.40968 0.41578
## SpeciesSisyrinchium_bellum -2.97710 0.53525
## SpeciesStachys_ajugoides_var_rigida -1.98667 0.45154
## Year2022 -1.04746 0.11953
## Year2023 0.20289 0.10706
## Spod_perc_survival:SpeciesArtemisia_californica -0.98841 0.89739
## Spod_perc_survival:SpeciesDiplacus_aurantiacus -0.34732 0.81767
## Spod_perc_survival:SpeciesEncelia_californica -1.16649 1.19042
## Spod_perc_survival:SpeciesEriogonum_fasciculatum 0.56948 0.84080
## Spod_perc_survival:SpeciesGrindelia_camporum -0.64164 0.93433
## Spod_perc_survival:SpeciesIsocoma_menziesii 0.30446 0.87739
## Spod_perc_survival:SpeciesMalacothamnus_fasciculatus 0.25489 0.99225
## Spod_perc_survival:SpeciesMalacothrix_saxatilis -0.26545 1.01887
## Spod_perc_survival:SpeciesMirabilis_laevis 2.89676 1.29187
## Spod_perc_survival:SpeciesSalvia_apiana -0.12024 0.82654
## Spod_perc_survival:SpeciesSalvia_mellifera -1.18806 0.91763
## Spod_perc_survival:SpeciesSisyrinchium_bellum 2.47432 1.27981
## Spod_perc_survival:SpeciesStachys_ajugoides_var_rigida -62.28671 9530.26704
## z value Pr(>|z|)
## (Intercept) 5.098 3.43e-07 ***
## Spod_perc_survival 0.040 0.9683
## SpeciesArtemisia_californica -1.167 0.2432
## SpeciesDiplacus_aurantiacus -0.430 0.6674
## SpeciesEncelia_californica -0.923 0.3560
## SpeciesEriogonum_fasciculatum -1.401 0.1612
## SpeciesGrindelia_camporum -1.232 0.2178
## SpeciesIsocoma_menziesii -1.155 0.2479
## SpeciesMalacothamnus_fasciculatus -1.986 0.0471 *
## SpeciesMalacothrix_saxatilis -2.604 0.0092 **
## SpeciesMirabilis_laevis -4.428 9.49e-06 ***
## SpeciesSalvia_apiana -1.338 0.1809
## SpeciesSalvia_mellifera -0.985 0.3245
## SpeciesSisyrinchium_bellum -5.562 2.67e-08 ***
## SpeciesStachys_ajugoides_var_rigida -4.400 1.08e-05 ***
## Year2022 -8.763 < 2e-16 ***
## Year2023 1.895 0.0581 .
## Spod_perc_survival:SpeciesArtemisia_californica -1.101 0.2707
## Spod_perc_survival:SpeciesDiplacus_aurantiacus -0.425 0.6710
## Spod_perc_survival:SpeciesEncelia_californica -0.980 0.3271
## Spod_perc_survival:SpeciesEriogonum_fasciculatum 0.677 0.4982
## Spod_perc_survival:SpeciesGrindelia_camporum -0.687 0.4922
## Spod_perc_survival:SpeciesIsocoma_menziesii 0.347 0.7286
## Spod_perc_survival:SpeciesMalacothamnus_fasciculatus 0.257 0.7973
## Spod_perc_survival:SpeciesMalacothrix_saxatilis -0.261 0.7944
## Spod_perc_survival:SpeciesMirabilis_laevis 2.242 0.0249 *
## Spod_perc_survival:SpeciesSalvia_apiana -0.145 0.8843
## Spod_perc_survival:SpeciesSalvia_mellifera -1.295 0.1954
## Spod_perc_survival:SpeciesSisyrinchium_bellum 1.933 0.0532 .
## Spod_perc_survival:SpeciesStachys_ajugoides_var_rigida -0.007 0.9948
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1## look at plot of model
plot_model(waspcount_surv, type = "pred", terms = c("Spod_perc_survival", "Species"), se - T,ci.lvl = .5)+
ggtitle("Predicted total wasp counts as a function of spodoptera survival")## Warning in RColorBrewer::brewer.pal(n, pal): n too large, allowed maximum for palette Set1 is 9
## Returning the palette you asked for with that many colors
TAKEAWAY Wasp count is significantly negatively correlated with spodoptera survival both for species by themselves and for the interaction between spodoptera survival * species.
2.3.7 log acarina count ~ Spodoptera survival
mitecount_surv <- glmmTMB(Abundance_Acarina ~ Spod_perc_survival * Species + Year + (1|Plot) + (1|Watering_Treatment) + (1|Plant), data = total_plant_order, family = poisson())## Warning in finalizeTMB(TMBStruc, obj, fit, h, data.tmb.old): Model convergence
## problem; singular convergence (7). See vignette('troubleshooting'),
## help('diagnose')##Test dispersion with Dharma package.
testDispersion(mitecount_surv)
##
## DHARMa nonparametric dispersion test via sd of residuals fitted vs.
## simulated
##
## data: simulationOutput
## dispersion = 0.99767, p-value = 0.728
## alternative hypothesis: two.sided## results indicate the data is not over- or uner-dispersed and we can stick with poisson and not have to switch to negative binomial.
Anova(mitecount_surv, type = 3)| Chisq | Df | Pr(>Chisq) |
|---|---|---|
| 3.37 | 1 | 0.0663 |
| 4.41 | 1 | 0.0356 |
| 156 | 13 | 1.2e-26 |
| 676 | 2 | 1.97e-147 |
| 53.4 | 13 | 7.8e-07 |
summary(mitecount_surv)## Family: poisson ( log )
## Formula:
## Abundance_Acarina ~ Spod_perc_survival * Species + Year + (1 |
## Plot) + (1 | Watering_Treatment) + (1 | Plant)
## Data: total_plant_order
##
## AIC BIC logLik deviance df.resid
## 2408.0 2553.3 -1171.0 2342.0 572
##
## Random effects:
##
## Conditional model:
## Groups Name Variance Std.Dev.
## Plot (Intercept) 4.086e-11 6.392e-06
## Watering_Treatment (Intercept) 2.476e-02 1.574e-01
## Plant (Intercept) 8.093e-01 8.996e-01
## Number of obs: 605, groups: Plot, 20; Watering_Treatment, 6; Plant, 246
##
## Conditional model:
## Estimate Std. Error
## (Intercept) -0.92023 0.50102
## Spod_perc_survival -1.76563 0.84030
## SpeciesArtemisia_californica 0.89055 0.53764
## SpeciesDiplacus_aurantiacus -0.56794 0.55791
## SpeciesEncelia_californica -0.39453 0.55455
## SpeciesEriogonum_fasciculatum -0.95874 0.60583
## SpeciesGrindelia_camporum 1.69807 0.57914
## SpeciesIsocoma_menziesii 1.24376 0.55633
## SpeciesMalacothamnus_fasciculatus 0.66565 0.56237
## SpeciesMalacothrix_saxatilis -0.94370 0.61282
## SpeciesMirabilis_laevis -2.66016 0.64375
## SpeciesSalvia_apiana 0.28116 0.57464
## SpeciesSalvia_mellifera 0.08955 0.55735
## SpeciesSisyrinchium_bellum -1.02989 0.62903
## SpeciesStachys_ajugoides_var_rigida -1.83316 0.66538
## Year2022 0.74908 0.11073
## Year2023 2.28523 0.10129
## Spod_perc_survival:SpeciesArtemisia_californica 0.82653 0.89065
## Spod_perc_survival:SpeciesDiplacus_aurantiacus 2.40910 0.97291
## Spod_perc_survival:SpeciesEncelia_californica 1.34269 1.26860
## Spod_perc_survival:SpeciesEriogonum_fasciculatum 3.28793 0.93820
## Spod_perc_survival:SpeciesGrindelia_camporum 0.16810 0.92209
## Spod_perc_survival:SpeciesIsocoma_menziesii 0.98017 0.91269
## Spod_perc_survival:SpeciesMalacothamnus_fasciculatus 1.90078 1.09013
## Spod_perc_survival:SpeciesMalacothrix_saxatilis 1.35927 1.22184
## Spod_perc_survival:SpeciesMirabilis_laevis 1.70137 2.57420
## Spod_perc_survival:SpeciesSalvia_apiana 1.91377 0.90651
## Spod_perc_survival:SpeciesSalvia_mellifera 2.17613 0.98754
## Spod_perc_survival:SpeciesSisyrinchium_bellum 3.73521 1.22307
## Spod_perc_survival:SpeciesStachys_ajugoides_var_rigida -57.21638 9084.50238
## z value Pr(>|z|)
## (Intercept) -1.837 0.066255 .
## Spod_perc_survival -2.101 0.035624 *
## SpeciesArtemisia_californica 1.656 0.097638 .
## SpeciesDiplacus_aurantiacus -1.018 0.308690
## SpeciesEncelia_californica -0.711 0.476805
## SpeciesEriogonum_fasciculatum -1.583 0.113531
## SpeciesGrindelia_camporum 2.932 0.003367 **
## SpeciesIsocoma_menziesii 2.236 0.025373 *
## SpeciesMalacothamnus_fasciculatus 1.184 0.236549
## SpeciesMalacothrix_saxatilis -1.540 0.123574
## SpeciesMirabilis_laevis -4.132 3.59e-05 ***
## SpeciesSalvia_apiana 0.489 0.624636
## SpeciesSalvia_mellifera 0.161 0.872345
## SpeciesSisyrinchium_bellum -1.637 0.101576
## SpeciesStachys_ajugoides_var_rigida -2.755 0.005868 **
## Year2022 6.765 1.33e-11 ***
## Year2023 22.562 < 2e-16 ***
## Spod_perc_survival:SpeciesArtemisia_californica 0.928 0.353399
## Spod_perc_survival:SpeciesDiplacus_aurantiacus 2.476 0.013279 *
## Spod_perc_survival:SpeciesEncelia_californica 1.058 0.289872
## Spod_perc_survival:SpeciesEriogonum_fasciculatum 3.505 0.000457 ***
## Spod_perc_survival:SpeciesGrindelia_camporum 0.182 0.855343
## Spod_perc_survival:SpeciesIsocoma_menziesii 1.074 0.282850
## Spod_perc_survival:SpeciesMalacothamnus_fasciculatus 1.744 0.081226 .
## Spod_perc_survival:SpeciesMalacothrix_saxatilis 1.112 0.265932
## Spod_perc_survival:SpeciesMirabilis_laevis 0.661 0.508654
## Spod_perc_survival:SpeciesSalvia_apiana 2.111 0.034760 *
## Spod_perc_survival:SpeciesSalvia_mellifera 2.204 0.027554 *
## Spod_perc_survival:SpeciesSisyrinchium_bellum 3.054 0.002258 **
## Spod_perc_survival:SpeciesStachys_ajugoides_var_rigida -0.006 0.994975
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1## look at plot of model
plot_model(mitecount_surv, type = "pred", terms = c("Spod_perc_survival", "Species"), se - T,ci.lvl = .5)+
ggtitle("Predicted total mite counts as a function of spodoptera survival")## Warning in RColorBrewer::brewer.pal(n, pal): n too large, allowed maximum for palette Set1 is 9
## Returning the palette you asked for with that many colors
TAKEAWAY Somewhat significant negative relationship between spodoptera survial and mite counts.
2.3.8 log thrip count ~ Spodoptera survival
thripcount_surv <- glmmTMB(Abundance_Thysanoptera ~ Spod_perc_survival * Species + Year + (1|Plot) + (1|Watering_Treatment) + (1|Plant), data = total_plant_order, family = nbinom2())## Warning in finalizeTMB(TMBStruc, obj, fit, h, data.tmb.old): Model convergence
## problem; singular convergence (7). See vignette('troubleshooting'),
## help('diagnose')##Test dispersion with Dharma package.
testDispersion(thripcount_surv)
##
## DHARMa nonparametric dispersion test via sd of residuals fitted vs.
## simulated
##
## data: simulationOutput
## dispersion = 1.1668, p-value = 0.536
## alternative hypothesis: two.sided## results indicate the data is over-dispersed and we need to switch to negative binomial.
Anova(thripcount_surv, type = 3)| Chisq | Df | Pr(>Chisq) |
|---|---|---|
| 15.8 | 1 | 7.21e-05 |
| 1.99 | 1 | 0.159 |
| 232 | 13 | 3.57e-42 |
| 2.22 | 2 | 0.33 |
| 20.6 | 13 | 0.0804 |
summary(thripcount_surv)## Family: nbinom2 ( log )
## Formula:
## Abundance_Thysanoptera ~ Spod_perc_survival * Species + Year +
## (1 | Plot) + (1 | Watering_Treatment) + (1 | Plant)
## Data: total_plant_order
##
## AIC BIC logLik deviance df.resid
## 3160.7 3310.6 -1546.4 3092.7 572
##
## Random effects:
##
## Conditional model:
## Groups Name Variance Std.Dev.
## Plot (Intercept) 7.401e-12 2.721e-06
## Watering_Treatment (Intercept) 5.372e-12 2.318e-06
## Plant (Intercept) 8.694e-11 9.324e-06
## Number of obs: 606, groups: Plot, 20; Watering_Treatment, 6; Plant, 246
##
## Dispersion parameter for nbinom2 family (): 0.342
##
## Conditional model:
## Estimate Std. Error
## (Intercept) 3.398e+00 8.562e-01
## Spod_perc_survival -2.195e+00 1.558e+00
## SpeciesArtemisia_californica -1.501e+00 9.267e-01
## SpeciesDiplacus_aurantiacus -1.826e+00 9.235e-01
## SpeciesEncelia_californica -2.129e+00 9.227e-01
## SpeciesEriogonum_fasciculatum -1.501e+00 1.022e+00
## SpeciesGrindelia_camporum 1.037e+00 1.015e+00
## SpeciesIsocoma_menziesii -7.413e-01 9.133e-01
## SpeciesMalacothamnus_fasciculatus 1.614e+00 9.127e-01
## SpeciesMalacothrix_saxatilis -1.393e+00 9.334e-01
## SpeciesMirabilis_laevis -2.981e+00 9.319e-01
## SpeciesSalvia_apiana -8.894e-03 9.331e-01
## SpeciesSalvia_mellifera -3.486e+00 9.322e-01
## SpeciesSisyrinchium_bellum -3.775e+00 9.533e-01
## SpeciesStachys_ajugoides_var_rigida -3.975e+00 9.591e-01
## Year2022 -9.372e-02 2.653e-01
## Year2023 -3.420e-01 2.528e-01
## Spod_perc_survival:SpeciesArtemisia_californica 2.339e+00 1.778e+00
## Spod_perc_survival:SpeciesDiplacus_aurantiacus 3.433e+00 1.707e+00
## Spod_perc_survival:SpeciesEncelia_californica 8.959e-01 2.250e+00
## Spod_perc_survival:SpeciesEriogonum_fasciculatum 3.409e+00 1.802e+00
## Spod_perc_survival:SpeciesGrindelia_camporum 7.572e-01 1.894e+00
## Spod_perc_survival:SpeciesIsocoma_menziesii 7.326e-01 1.850e+00
## Spod_perc_survival:SpeciesMalacothamnus_fasciculatus 2.261e+00 2.035e+00
## Spod_perc_survival:SpeciesMalacothrix_saxatilis 7.634e-01 1.869e+00
## Spod_perc_survival:SpeciesMirabilis_laevis -5.020e-01 2.626e+00
## Spod_perc_survival:SpeciesSalvia_apiana 7.399e-01 1.745e+00
## Spod_perc_survival:SpeciesSalvia_mellifera 2.691e+00 1.872e+00
## Spod_perc_survival:SpeciesSisyrinchium_bellum 4.309e+00 2.066e+00
## Spod_perc_survival:SpeciesStachys_ajugoides_var_rigida -6.014e+01 7.813e+03
## z value Pr(>|z|)
## (Intercept) 3.969 7.21e-05 ***
## Spod_perc_survival -1.409 0.158780
## SpeciesArtemisia_californica -1.619 0.105343
## SpeciesDiplacus_aurantiacus -1.978 0.047963 *
## SpeciesEncelia_californica -2.308 0.021020 *
## SpeciesEriogonum_fasciculatum -1.469 0.141920
## SpeciesGrindelia_camporum 1.022 0.306879
## SpeciesIsocoma_menziesii -0.812 0.416995
## SpeciesMalacothamnus_fasciculatus 1.768 0.076995 .
## SpeciesMalacothrix_saxatilis -1.492 0.135598
## SpeciesMirabilis_laevis -3.199 0.001380 **
## SpeciesSalvia_apiana -0.010 0.992395
## SpeciesSalvia_mellifera -3.739 0.000184 ***
## SpeciesSisyrinchium_bellum -3.960 7.49e-05 ***
## SpeciesStachys_ajugoides_var_rigida -4.144 3.41e-05 ***
## Year2022 -0.353 0.723864
## Year2023 -1.353 0.176122
## Spod_perc_survival:SpeciesArtemisia_californica 1.315 0.188405
## Spod_perc_survival:SpeciesDiplacus_aurantiacus 2.011 0.044360 *
## Spod_perc_survival:SpeciesEncelia_californica 0.398 0.690515
## Spod_perc_survival:SpeciesEriogonum_fasciculatum 1.892 0.058521 .
## Spod_perc_survival:SpeciesGrindelia_camporum 0.400 0.689243
## Spod_perc_survival:SpeciesIsocoma_menziesii 0.396 0.692097
## Spod_perc_survival:SpeciesMalacothamnus_fasciculatus 1.112 0.266333
## Spod_perc_survival:SpeciesMalacothrix_saxatilis 0.408 0.682984
## Spod_perc_survival:SpeciesMirabilis_laevis -0.191 0.848397
## Spod_perc_survival:SpeciesSalvia_apiana 0.424 0.671550
## Spod_perc_survival:SpeciesSalvia_mellifera 1.438 0.150517
## Spod_perc_survival:SpeciesSisyrinchium_bellum 2.085 0.037061 *
## Spod_perc_survival:SpeciesStachys_ajugoides_var_rigida -0.008 0.993859
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1## look at plot of model
plot_model(thripcount_surv, type = "pred", terms = c("Spod_perc_survival", "Species"), se - T,ci.lvl = .5)+
ggtitle("Predicted total thrip counts as a function of spodoptera survival")## Warning in RColorBrewer::brewer.pal(n, pal): n too large, allowed maximum for palette Set1 is 9
## Returning the palette you asked for with that many colors
TAKEAWAY Nonsignificant relationship between spodoptera survival and thrip counts.
2.3.9 log Coleoptera count ~ Spodoptera survival
coleocount_surv <- glmmTMB(Abundance_Coleoptera ~ Spod_perc_survival * Species + Year + (1|Plot) + (1|Watering_Treatment) + (1|Plant), data = total_plant_order, family = nbinom2())## Warning in finalizeTMB(TMBStruc, obj, fit, h, data.tmb.old): Model convergence
## problem; non-positive-definite Hessian matrix. See vignette('troubleshooting')## Warning in finalizeTMB(TMBStruc, obj, fit, h, data.tmb.old): Model convergence
## problem; singular convergence (7). See vignette('troubleshooting'),
## help('diagnose')##Test dispersion with Dharma package.
testDispersion(coleocount_surv)
##
## DHARMa nonparametric dispersion test via sd of residuals fitted vs.
## simulated
##
## data: simulationOutput
## dispersion = 1.0934, p-value = 0.528
## alternative hypothesis: two.sided## results indicate the data is over-dispersed and we need to switch to negative binomial.
Anova(coleocount_surv, type = 3)| Chisq | Df | Pr(>Chisq) |
|---|---|---|
| 5.11 | 1 | 0.0238 |
| 0.276 | 1 | 0.599 |
| 219 | 13 | 2.1e-39 |
| 4.83 | 2 | 0.0894 |
| 19.7 | 13 | 0.102 |
summary(coleocount_surv)## Family: nbinom2 ( log )
## Formula: Abundance_Coleoptera ~ Spod_perc_survival * Species + Year +
## (1 | Plot) + (1 | Watering_Treatment) + (1 | Plant)
## Data: total_plant_order
##
## AIC BIC logLik deviance df.resid
## NA NA NA NA 572
##
## Random effects:
##
## Conditional model:
## Groups Name Variance Std.Dev.
## Plot (Intercept) 1.840e-26 1.356e-13
## Watering_Treatment (Intercept) 7.829e-29 8.848e-15
## Plant (Intercept) 5.274e-23 7.262e-12
## Number of obs: 606, groups: Plot, 20; Watering_Treatment, 6; Plant, 246
##
## Dispersion parameter for nbinom2 family (): 1.05
##
## Conditional model:
## Estimate Std. Error
## (Intercept) 1.056e+00 4.671e-01
## Spod_perc_survival -4.329e-01 8.243e-01
## SpeciesArtemisia_californica -5.939e-01 5.104e-01
## SpeciesDiplacus_aurantiacus 1.057e+00 4.952e-01
## SpeciesEncelia_californica 1.877e+00 4.949e-01
## SpeciesEriogonum_fasciculatum 6.282e-01 5.417e-01
## SpeciesGrindelia_camporum -6.821e-03 5.817e-01
## SpeciesIsocoma_menziesii -1.926e-01 5.149e-01
## SpeciesMalacothamnus_fasciculatus 1.543e+00 5.029e-01
## SpeciesMalacothrix_saxatilis 7.848e-01 5.165e-01
## SpeciesMirabilis_laevis 1.554e-02 4.984e-01
## SpeciesSalvia_apiana 2.192e-01 5.224e-01
## SpeciesSalvia_mellifera 1.494e-01 5.035e-01
## SpeciesSisyrinchium_bellum -2.963e-01 5.268e-01
## SpeciesStachys_ajugoides_var_rigida -9.504e-01 5.280e-01
## Year2022 -2.335e-01 1.229e-01
## Year2023 1.336e-02 1.286e-01
## Spod_perc_survival:SpeciesArtemisia_californica 1.736e-01 1.029e+00
## Spod_perc_survival:SpeciesDiplacus_aurantiacus 4.931e-01 9.373e-01
## Spod_perc_survival:SpeciesEncelia_californica -1.335e+00 1.206e+00
## Spod_perc_survival:SpeciesEriogonum_fasciculatum 4.124e-01 9.458e-01
## Spod_perc_survival:SpeciesGrindelia_camporum -5.213e-01 1.084e+00
## Spod_perc_survival:SpeciesIsocoma_menziesii 3.827e-01 1.065e+00
## Spod_perc_survival:SpeciesMalacothamnus_fasciculatus 8.934e-01 1.043e+00
## Spod_perc_survival:SpeciesMalacothrix_saxatilis -8.500e-01 1.062e+00
## Spod_perc_survival:SpeciesMirabilis_laevis 3.036e+00 1.314e+00
## Spod_perc_survival:SpeciesSalvia_apiana 1.534e-01 9.530e-01
## Spod_perc_survival:SpeciesSalvia_mellifera 5.060e-01 1.020e+00
## Spod_perc_survival:SpeciesSisyrinchium_bellum 1.914e+00 1.184e+00
## Spod_perc_survival:SpeciesStachys_ajugoides_var_rigida -6.402e+01 7.762e+03
## z value Pr(>|z|)
## (Intercept) 2.261 0.023781 *
## Spod_perc_survival -0.525 0.599429
## SpeciesArtemisia_californica -1.164 0.244580
## SpeciesDiplacus_aurantiacus 2.134 0.032866 *
## SpeciesEncelia_californica 3.793 0.000149 ***
## SpeciesEriogonum_fasciculatum 1.160 0.246241
## SpeciesGrindelia_camporum -0.012 0.990644
## SpeciesIsocoma_menziesii -0.374 0.708338
## SpeciesMalacothamnus_fasciculatus 3.069 0.002147 **
## SpeciesMalacothrix_saxatilis 1.520 0.128621
## SpeciesMirabilis_laevis 0.031 0.975119
## SpeciesSalvia_apiana 0.420 0.674830
## SpeciesSalvia_mellifera 0.297 0.766706
## SpeciesSisyrinchium_bellum -0.562 0.573858
## SpeciesStachys_ajugoides_var_rigida -1.800 0.071888 .
## Year2022 -1.900 0.057374 .
## Year2023 0.104 0.917244
## Spod_perc_survival:SpeciesArtemisia_californica 0.169 0.866069
## Spod_perc_survival:SpeciesDiplacus_aurantiacus 0.526 0.598805
## Spod_perc_survival:SpeciesEncelia_californica -1.107 0.268155
## Spod_perc_survival:SpeciesEriogonum_fasciculatum 0.436 0.662783
## Spod_perc_survival:SpeciesGrindelia_camporum -0.481 0.630554
## Spod_perc_survival:SpeciesIsocoma_menziesii 0.359 0.719307
## Spod_perc_survival:SpeciesMalacothamnus_fasciculatus 0.856 0.391824
## Spod_perc_survival:SpeciesMalacothrix_saxatilis -0.801 0.423376
## Spod_perc_survival:SpeciesMirabilis_laevis 2.310 0.020911 *
## Spod_perc_survival:SpeciesSalvia_apiana 0.161 0.872103
## Spod_perc_survival:SpeciesSalvia_mellifera 0.496 0.619767
## Spod_perc_survival:SpeciesSisyrinchium_bellum 1.617 0.105948
## Spod_perc_survival:SpeciesStachys_ajugoides_var_rigida -0.008 0.993420
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1## look at plot of model
plot_model(coleocount_surv, type = "pred", terms = c("Spod_perc_survival", "Species"), se - T,ci.lvl = .5)+
ggtitle("Predicted total beetle counts as a function of spodoptera survival")## Warning in RColorBrewer::brewer.pal(n, pal): n too large, allowed maximum for palette Set1 is 9
## Returning the palette you asked for with that many colors
TAKEAWAY Coleoptera show no significant relationship with spodoptera survival though certain species by themselves do.
2.3.10 log Diptera count ~ Spodoptera survival
diptcount_surv <- glmmTMB(Abundance_Diptera ~ Spod_perc_survival * Species + Year + (1|Plot) + (1|Watering_Treatment) + (1|Plant), data = total_plant_order, family = poisson())
##Test dispersion with Dharma package.
testDispersion(diptcount_surv)
##
## DHARMa nonparametric dispersion test via sd of residuals fitted vs.
## simulated
##
## data: simulationOutput
## dispersion = 0.91582, p-value = 0.984
## alternative hypothesis: two.sided## results indicate the data is not over- or uner-dispersed and we can stick with poisson and not have to switch to negative binomial.
Anova(diptcount_surv, type = 3)| Chisq | Df | Pr(>Chisq) |
|---|---|---|
| 8.68 | 1 | 0.00322 |
| 3.16 | 1 | 0.0757 |
| 101 | 13 | 1.17e-15 |
| 788 | 2 | 6.89e-172 |
| 72.8 | 13 | 2.45e-10 |
summary(diptcount_surv)## Family: poisson ( log )
## Formula:
## Abundance_Diptera ~ Spod_perc_survival * Species + Year + (1 |
## Plot) + (1 | Watering_Treatment) + (1 | Plant)
## Data: total_plant_order
##
## AIC BIC logLik deviance df.resid
## 2183.9 2329.3 -1058.9 2117.9 573
##
## Random effects:
##
## Conditional model:
## Groups Name Variance Std.Dev.
## Plot (Intercept) 0.008713 0.09334
## Watering_Treatment (Intercept) 0.027930 0.16712
## Plant (Intercept) 0.412323 0.64212
## Number of obs: 606, groups: Plot, 20; Watering_Treatment, 6; Plant, 246
##
## Conditional model:
## Estimate Std. Error
## (Intercept) -1.54986 0.52616
## Spod_perc_survival 1.34128 0.75511
## SpeciesArtemisia_californica 1.09120 0.55077
## SpeciesDiplacus_aurantiacus 0.97952 0.55504
## SpeciesEncelia_californica 1.11534 0.55413
## SpeciesEriogonum_fasciculatum 2.13829 0.56432
## SpeciesGrindelia_camporum 2.52202 0.58696
## SpeciesIsocoma_menziesii 0.93422 0.57929
## SpeciesMalacothamnus_fasciculatus 1.69206 0.56382
## SpeciesMalacothrix_saxatilis 0.10161 0.61090
## SpeciesMirabilis_laevis 1.01419 0.55354
## SpeciesSalvia_apiana 0.44260 0.59708
## SpeciesSalvia_mellifera 1.50871 0.55557
## SpeciesSisyrinchium_bellum -0.14703 0.63684
## SpeciesStachys_ajugoides_var_rigida 0.29456 0.59025
## Year2022 0.11462 0.11288
## Year2023 1.96076 0.08862
## Spod_perc_survival:SpeciesArtemisia_californica -2.59281 1.00776
## Spod_perc_survival:SpeciesDiplacus_aurantiacus -0.13852 0.81660
## Spod_perc_survival:SpeciesEncelia_californica -2.96981 1.36329
## Spod_perc_survival:SpeciesEriogonum_fasciculatum -1.67950 0.80135
## Spod_perc_survival:SpeciesGrindelia_camporum -4.83490 1.02518
## Spod_perc_survival:SpeciesIsocoma_menziesii -1.17750 0.90204
## Spod_perc_survival:SpeciesMalacothamnus_fasciculatus -2.08755 0.99980
## Spod_perc_survival:SpeciesMalacothrix_saxatilis 0.11751 1.04989
## Spod_perc_survival:SpeciesMirabilis_laevis -1.79471 1.16641
## Spod_perc_survival:SpeciesSalvia_apiana 0.17387 0.83866
## Spod_perc_survival:SpeciesSalvia_mellifera -0.55160 0.84860
## Spod_perc_survival:SpeciesSisyrinchium_bellum -1.04197 1.27768
## Spod_perc_survival:SpeciesStachys_ajugoides_var_rigida 0.17648 4.29652
## z value Pr(>|z|)
## (Intercept) -2.946 0.003223 **
## Spod_perc_survival 1.776 0.075690 .
## SpeciesArtemisia_californica 1.981 0.047567 *
## SpeciesDiplacus_aurantiacus 1.765 0.077600 .
## SpeciesEncelia_californica 2.013 0.044139 *
## SpeciesEriogonum_fasciculatum 3.789 0.000151 ***
## SpeciesGrindelia_camporum 4.297 1.73e-05 ***
## SpeciesIsocoma_menziesii 1.613 0.106806
## SpeciesMalacothamnus_fasciculatus 3.001 0.002691 **
## SpeciesMalacothrix_saxatilis 0.166 0.867903
## SpeciesMirabilis_laevis 1.832 0.066926 .
## SpeciesSalvia_apiana 0.741 0.458525
## SpeciesSalvia_mellifera 2.716 0.006615 **
## SpeciesSisyrinchium_bellum -0.231 0.817411
## SpeciesStachys_ajugoides_var_rigida 0.499 0.617757
## Year2022 1.015 0.309911
## Year2023 22.126 < 2e-16 ***
## Spod_perc_survival:SpeciesArtemisia_californica -2.573 0.010087 *
## Spod_perc_survival:SpeciesDiplacus_aurantiacus -0.170 0.865305
## Spod_perc_survival:SpeciesEncelia_californica -2.178 0.029375 *
## Spod_perc_survival:SpeciesEriogonum_fasciculatum -2.096 0.036096 *
## Spod_perc_survival:SpeciesGrindelia_camporum -4.716 2.40e-06 ***
## Spod_perc_survival:SpeciesIsocoma_menziesii -1.305 0.191767
## Spod_perc_survival:SpeciesMalacothamnus_fasciculatus -2.088 0.036801 *
## Spod_perc_survival:SpeciesMalacothrix_saxatilis 0.112 0.910881
## Spod_perc_survival:SpeciesMirabilis_laevis -1.539 0.123887
## Spod_perc_survival:SpeciesSalvia_apiana 0.207 0.835756
## Spod_perc_survival:SpeciesSalvia_mellifera -0.650 0.515682
## Spod_perc_survival:SpeciesSisyrinchium_bellum -0.816 0.414775
## Spod_perc_survival:SpeciesStachys_ajugoides_var_rigida 0.041 0.967236
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1## look at plot of model
plot_model(diptcount_surv, type = "pred", terms = c("Spod_perc_survival", "Species"), se - T,ci.lvl = .5)+
ggtitle("Predicted total fly counts as a function of spodoptera survival")## Warning in RColorBrewer::brewer.pal(n, pal): n too large, allowed maximum for palette Set1 is 9
## Returning the palette you asked for with that many colors
TAKEAWAY No significant relationship here- effect seems to be neutral.
2.3.11 log Aranea count ~ Spodoptera survival
aracount_surv <- glmmTMB(Abundance_Aranea ~ Spod_perc_survival * Species + Year + (1|Plot) + (1|Watering_Treatment) + (1|Plant), data = total_plant_order, family = nbinom2())## Warning in finalizeTMB(TMBStruc, obj, fit, h, data.tmb.old): Model convergence
## problem; singular convergence (7). See vignette('troubleshooting'),
## help('diagnose')##Test dispersion with Dharma package.
testDispersion(aracount_surv)
##
## DHARMa nonparametric dispersion test via sd of residuals fitted vs.
## simulated
##
## data: simulationOutput
## dispersion = 1.0938, p-value = 0.528
## alternative hypothesis: two.sided## results indicate the data is under-dispersed and we need to switch to negative binomial.
Anova(aracount_surv, type = 3)| Chisq | Df | Pr(>Chisq) |
|---|---|---|
| 0.00611 | 1 | 0.938 |
| 1.18 | 1 | 0.277 |
| 54.5 | 13 | 4.93e-07 |
| 17.5 | 2 | 0.000156 |
| 12.2 | 13 | 0.508 |
summary(aracount_surv)## Family: nbinom2 ( log )
## Formula:
## Abundance_Aranea ~ Spod_perc_survival * Species + Year + (1 |
## Plot) + (1 | Watering_Treatment) + (1 | Plant)
## Data: total_plant_order
##
## AIC BIC logLik deviance df.resid
## 1686.6 1836.4 -809.3 1618.6 571
##
## Random effects:
##
## Conditional model:
## Groups Name Variance Std.Dev.
## Plot (Intercept) 5.404e-16 2.325e-08
## Watering_Treatment (Intercept) 5.008e-03 7.076e-02
## Plant (Intercept) 4.400e-02 2.098e-01
## Number of obs: 605, groups: Plot, 20; Watering_Treatment, 6; Plant, 246
##
## Dispersion parameter for nbinom2 family (): 1.3
##
## Conditional model:
## Estimate Std. Error
## (Intercept) -4.432e-02 5.667e-01
## Spod_perc_survival -1.159e+00 1.067e+00
## SpeciesArtemisia_californica 5.170e-01 6.020e-01
## SpeciesDiplacus_aurantiacus 9.306e-01 5.942e-01
## SpeciesEncelia_californica -3.870e-02 6.086e-01
## SpeciesEriogonum_fasciculatum 4.730e-01 6.576e-01
## SpeciesGrindelia_camporum 5.910e-02 6.953e-01
## SpeciesIsocoma_menziesii 1.174e+00 6.015e-01
## SpeciesMalacothamnus_fasciculatus 5.833e-02 6.191e-01
## SpeciesMalacothrix_saxatilis 4.355e-01 6.258e-01
## SpeciesMirabilis_laevis -3.632e-01 6.170e-01
## SpeciesSalvia_apiana -9.791e-02 6.440e-01
## SpeciesSalvia_mellifera 3.491e-01 6.065e-01
## SpeciesSisyrinchium_bellum -3.467e-01 6.426e-01
## SpeciesStachys_ajugoides_var_rigida -7.616e-01 6.557e-01
## Year2022 -5.610e-01 1.458e-01
## Year2023 -4.477e-01 1.498e-01
## Spod_perc_survival:SpeciesArtemisia_californica 6.584e-01 1.252e+00
## Spod_perc_survival:SpeciesDiplacus_aurantiacus 1.515e+00 1.182e+00
## Spod_perc_survival:SpeciesEncelia_californica 5.952e-01 1.713e+00
## Spod_perc_survival:SpeciesEriogonum_fasciculatum 9.156e-01 1.225e+00
## Spod_perc_survival:SpeciesGrindelia_camporum 1.443e+00 1.323e+00
## Spod_perc_survival:SpeciesIsocoma_menziesii 1.589e+00 1.247e+00
## Spod_perc_survival:SpeciesMalacothamnus_fasciculatus 1.462e+00 1.391e+00
## Spod_perc_survival:SpeciesMalacothrix_saxatilis -4.977e-04 1.421e+00
## Spod_perc_survival:SpeciesMirabilis_laevis 3.717e+00 1.643e+00
## Spod_perc_survival:SpeciesSalvia_apiana 1.685e+00 1.228e+00
## Spod_perc_survival:SpeciesSalvia_mellifera 1.741e+00 1.281e+00
## Spod_perc_survival:SpeciesSisyrinchium_bellum 2.681e+00 1.391e+00
## Spod_perc_survival:SpeciesStachys_ajugoides_var_rigida -6.011e+01 9.467e+03
## z value Pr(>|z|)
## (Intercept) -0.078 0.937672
## Spod_perc_survival -1.087 0.277218
## SpeciesArtemisia_californica 0.859 0.390505
## SpeciesDiplacus_aurantiacus 1.566 0.117331
## SpeciesEncelia_californica -0.064 0.949296
## SpeciesEriogonum_fasciculatum 0.719 0.471962
## SpeciesGrindelia_camporum 0.085 0.932265
## SpeciesIsocoma_menziesii 1.952 0.050978 .
## SpeciesMalacothamnus_fasciculatus 0.094 0.924934
## SpeciesMalacothrix_saxatilis 0.696 0.486524
## SpeciesMirabilis_laevis -0.589 0.556149
## SpeciesSalvia_apiana -0.152 0.879167
## SpeciesSalvia_mellifera 0.576 0.564846
## SpeciesSisyrinchium_bellum -0.540 0.589517
## SpeciesStachys_ajugoides_var_rigida -1.161 0.245479
## Year2022 -3.849 0.000119 ***
## Year2023 -2.989 0.002797 **
## Spod_perc_survival:SpeciesArtemisia_californica 0.526 0.598836
## Spod_perc_survival:SpeciesDiplacus_aurantiacus 1.282 0.199979
## Spod_perc_survival:SpeciesEncelia_californica 0.348 0.728207
## Spod_perc_survival:SpeciesEriogonum_fasciculatum 0.747 0.454767
## Spod_perc_survival:SpeciesGrindelia_camporum 1.090 0.275656
## Spod_perc_survival:SpeciesIsocoma_menziesii 1.274 0.202771
## Spod_perc_survival:SpeciesMalacothamnus_fasciculatus 1.051 0.293214
## Spod_perc_survival:SpeciesMalacothrix_saxatilis 0.000 0.999720
## Spod_perc_survival:SpeciesMirabilis_laevis 2.262 0.023713 *
## Spod_perc_survival:SpeciesSalvia_apiana 1.372 0.170192
## Spod_perc_survival:SpeciesSalvia_mellifera 1.360 0.173948
## Spod_perc_survival:SpeciesSisyrinchium_bellum 1.927 0.053953 .
## Spod_perc_survival:SpeciesStachys_ajugoides_var_rigida -0.006 0.994934
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1## look at plot of model
plot_model(aracount_surv, type = "pred", terms = c("Spod_perc_survival", "Species"), se - T,ci.lvl = .5)+
ggtitle("Predicted total Aranea counts as a function of spodoptera survival")## Warning in RColorBrewer::brewer.pal(n, pal): n too large, allowed maximum for palette Set1 is 9
## Returning the palette you asked for with that many colors
TAKEAWAY Similar to diptera. Not much to see here.
Next I will just look at 2023 data and focus on the correlation between spodoptera survival and arthropod mass since I only have mass data for 2023. I will run the analyses both in terms of feeding guild and order.
2.4.1 Plot histograms to visualize mass data (feeding guilds)
## Plot histogram of mass to determine if poisson transformation necessary. Only focus on total, herbivores and predators
hist(total_plant$Arth_Mass_Total_mg)
hist(total_plant$Mass_Herb_mg)
hist(total_plant$Mass_Pred_mg)
## Looks like data is non-normal, however poisson transformation only works with integers and not decimals. I will run the model as gaussian and look at residuals.
## Create a new data set that only includes 2023 data since that is what I will be comparing mass to.
selected_years <- c(2023)
plant_2023 <- total_plant[total_plant$Year %in% selected_years,]
plant_2023## # A tibble: 280 × 39
## # Groups: Year [1]
## Plant Species Watering_Treatment Plot Year Spod_perc_survival
## <fct> <fct> <fct> <fct> <fct> <dbl>
## 1 1 Acmispon_glaber 1 1 2023 NA
## 2 2 Acmispon_glaber 1 2 2023 NA
## 3 3 Acmispon_glaber 1 3 2023 NA
## 4 4 Acmispon_glaber 1 4 2023 NA
## 5 5 Acmispon_glaber 2 5 2023 NA
## 6 6 Acmispon_glaber 2 6 2023 NA
## 7 7 Acmispon_glaber 2 7 2023 NA
## 8 8 Acmispon_glaber 3 8 2023 0
## 9 9 Acmispon_glaber 3 9 2023 NA
## 10 10 Acmispon_glaber 3 10 2023 NA
## # ℹ 270 more rows
## # ℹ 33 more variables: Avg_spod_mass_g <dbl>, Mass_Paly_mg <dbl>,
## # Abundance_Paly <int>, Mass_Detr_mg <dbl>, Abundance_Detr <int>,
## # Mass_Omni_mg <dbl>, Abundance_Omni <int>, Mass_Pred_mg <dbl>,
## # Abundance_Pred <int>, Mass_Herb_mg <dbl>, Abundance_Herb <int>,
## # Abundance_Total <int>, Arth_Mass_Total_mg <dbl>, Dry_Plant_Biomass_g <dbl>,
## # Dry_Biomass_Sampled_g <dbl>, Plot_X <int>, Plot_Y <int>, …TAKEAWAY Poisson wont work because of non-integer values. Stick with gaussian and look at residuals.
2.4.2 total mass ~ Spodoptera survival NOTE remove “year” from model as there is only one year where I have mass data.
## Run glmmTMB model for Total mass ~ Spodoptera percent survival. Family = gaussian.
arthmass_surv <- glmmTMB(Arth_Mass_Total_mg ~ Spod_perc_survival * Species + (1|Plot) + (1|Watering_Treatment) + (1|Plant), data = total_plant, family = gaussian())## dropping columns from rank-deficient conditional model: Spod_perc_survival:SpeciesGrindelia_camporum, Spod_perc_survival:SpeciesSisyrinchium_bellum, Spod_perc_survival:SpeciesStachys_ajugoides_var_rigida## Warning in finalizeTMB(TMBStruc, obj, fit, h, data.tmb.old): Model convergence
## problem; non-positive-definite Hessian matrix. See vignette('troubleshooting')##Test dispersion with Dharma package.
testDispersion(arthmass_surv)
##
## DHARMa nonparametric dispersion test via sd of residuals fitted vs.
## simulated
##
## data: simulationOutput
## dispersion = 1.0037, p-value = 0.952
## alternative hypothesis: two.sided## results indicate the data is not over- or under-dispersed and we can stick with gaussian
Anova(arthmass_surv, type = 2)| Chisq | Df | Pr(>Chisq) |
|---|---|---|
| 0.0367 | 1 | 0.848 |
| 55.2 | 13 | 3.76e-07 |
| 2.85 | 10 | 0.985 |
summary(arthmass_surv)## Family: gaussian ( identity )
## Formula:
## Arth_Mass_Total_mg ~ Spod_perc_survival * Species + (1 | Plot) +
## (1 | Watering_Treatment) + (1 | Plant)
## Data: total_plant
##
## AIC BIC logLik deviance df.resid
## NA NA NA NA 141
##
## Random effects:
##
## Conditional model:
## Groups Name Variance Std.Dev.
## Plot (Intercept) 5.303e-05 7.282e-03
## Watering_Treatment (Intercept) 2.080e-05 4.561e-03
## Plant (Intercept) 1.808e+04 1.345e+02
## Residual 3.245e-01 5.697e-01
## Number of obs: 170, groups: Plot, 20; Watering_Treatment, 6; Plant, 170
##
## Dispersion estimate for gaussian family (sigma^2): 0.325
##
## Conditional model:
## Estimate Std. Error
## (Intercept) 123.61 95.08
## Spod_perc_survival -3.11 174.51
## SpeciesArtemisia_californica -48.24 100.22
## SpeciesDiplacus_aurantiacus -105.60 102.03
## SpeciesEncelia_californica -81.03 101.64
## SpeciesEriogonum_fasciculatum -110.74 117.10
## SpeciesGrindelia_camporum -114.92 239.94
## SpeciesIsocoma_menziesii -91.51 110.01
## SpeciesMalacothamnus_fasciculatus -79.26 107.12
## SpeciesMalacothrix_saxatilis -19.32 107.12
## SpeciesMirabilis_laevis 178.38 100.85
## SpeciesSalvia_apiana -107.36 119.21
## SpeciesSalvia_mellifera -89.79 104.01
## SpeciesSisyrinchium_bellum -118.04 111.22
## SpeciesStachys_ajugoides_var_rigida -113.56 106.30
## Spod_perc_survival:SpeciesArtemisia_californica -78.97 437.23
## Spod_perc_survival:SpeciesDiplacus_aurantiacus 22.97 208.81
## Spod_perc_survival:SpeciesEncelia_californica -90.00 384.11
## Spod_perc_survival:SpeciesEriogonum_fasciculatum 48.07 206.50
## Spod_perc_survival:SpeciesGrindelia_camporum NA NA
## Spod_perc_survival:SpeciesIsocoma_menziesii -19.39 210.83
## Spod_perc_survival:SpeciesMalacothamnus_fasciculatus -10.43 252.65
## Spod_perc_survival:SpeciesMalacothrix_saxatilis -92.77 241.39
## Spod_perc_survival:SpeciesMirabilis_laevis -680.79 581.21
## Spod_perc_survival:SpeciesSalvia_apiana 12.21 202.20
## Spod_perc_survival:SpeciesSalvia_mellifera 93.40 206.50
## Spod_perc_survival:SpeciesSisyrinchium_bellum NA NA
## Spod_perc_survival:SpeciesStachys_ajugoides_var_rigida NA NA
## z value Pr(>|z|)
## (Intercept) 1.300 0.1936
## Spod_perc_survival -0.018 0.9858
## SpeciesArtemisia_californica -0.481 0.6302
## SpeciesDiplacus_aurantiacus -1.035 0.3006
## SpeciesEncelia_californica -0.797 0.4253
## SpeciesEriogonum_fasciculatum -0.946 0.3443
## SpeciesGrindelia_camporum -0.479 0.6320
## SpeciesIsocoma_menziesii -0.832 0.4055
## SpeciesMalacothamnus_fasciculatus -0.740 0.4593
## SpeciesMalacothrix_saxatilis -0.180 0.8568
## SpeciesMirabilis_laevis 1.769 0.0769 .
## SpeciesSalvia_apiana -0.901 0.3678
## SpeciesSalvia_mellifera -0.863 0.3880
## SpeciesSisyrinchium_bellum -1.061 0.2886
## SpeciesStachys_ajugoides_var_rigida -1.068 0.2854
## Spod_perc_survival:SpeciesArtemisia_californica -0.181 0.8567
## Spod_perc_survival:SpeciesDiplacus_aurantiacus 0.110 0.9124
## Spod_perc_survival:SpeciesEncelia_californica -0.234 0.8147
## Spod_perc_survival:SpeciesEriogonum_fasciculatum 0.233 0.8159
## Spod_perc_survival:SpeciesGrindelia_camporum NA NA
## Spod_perc_survival:SpeciesIsocoma_menziesii -0.092 0.9267
## Spod_perc_survival:SpeciesMalacothamnus_fasciculatus -0.041 0.9671
## Spod_perc_survival:SpeciesMalacothrix_saxatilis -0.384 0.7007
## Spod_perc_survival:SpeciesMirabilis_laevis -1.171 0.2415
## Spod_perc_survival:SpeciesSalvia_apiana 0.060 0.9519
## Spod_perc_survival:SpeciesSalvia_mellifera 0.452 0.6511
## Spod_perc_survival:SpeciesSisyrinchium_bellum NA NA
## Spod_perc_survival:SpeciesStachys_ajugoides_var_rigida NA NA
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1## look at plot of model
plot_model(arthmass_surv, type = "pred", terms = c("Spod_perc_survival", "Species"), se - T,ci.lvl = .5)+
ggtitle("Predicted total arthropod mass as a function of spodoptera survival")## Warning in RColorBrewer::brewer.pal(n, pal): n too large, allowed maximum for palette Set1 is 9
## Returning the palette you asked for with that many colors
TAKEAWAY Counts go down as a function of increasing spodoptera survival however, mass stays the same. One way to interpret this is that there are less arthropods but they are bigger. I think…
2.4.3 herbivore mass ~ Spodoptera survival
## Run glmmTMB model for herbivore mass ~ Spodoptera percent survival. Family = gaussian.
herbmass_surv <- glmmTMB(Mass_Herb_mg ~ Spod_perc_survival * Species + (1|Plot) + (1|Watering_Treatment) + (1|Plant), data = total_plant, family = gaussian())## dropping columns from rank-deficient conditional model: Spod_perc_survival:SpeciesGrindelia_camporum, Spod_perc_survival:SpeciesSisyrinchium_bellum, Spod_perc_survival:SpeciesStachys_ajugoides_var_rigida##Test dispersion with Dharma package.
testDispersion(herbmass_surv)
##
## DHARMa nonparametric dispersion test via sd of residuals fitted vs.
## simulated
##
## data: simulationOutput
## dispersion = 1.0037, p-value = 0.952
## alternative hypothesis: two.sided## results indicate the data is not over- or under-dispersed and we can stick with gaussian
Anova(herbmass_surv, type = 2)| Chisq | Df | Pr(>Chisq) |
|---|---|---|
| 7.23e-06 | 1 | 0.998 |
| 42.5 | 13 | 5.37e-05 |
| 1.87 | 10 | 0.997 |
summary(herbmass_surv)## Family: gaussian ( identity )
## Formula:
## Mass_Herb_mg ~ Spod_perc_survival * Species + (1 | Plot) + (1 |
## Watering_Treatment) + (1 | Plant)
## Data: total_plant
##
## AIC BIC logLik deviance df.resid
## 2199.0 2289.9 -1070.5 2141.0 141
##
## Random effects:
##
## Conditional model:
## Groups Name Variance Std.Dev.
## Plot (Intercept) 3.122e-04 1.767e-02
## Watering_Treatment (Intercept) 6.888e-06 2.624e-03
## Plant (Intercept) 1.726e+04 1.314e+02
## Residual 1.564e-01 3.955e-01
## Number of obs: 170, groups: Plot, 20; Watering_Treatment, 6; Plant, 170
##
## Dispersion estimate for gaussian family (sigma^2): 0.156
##
## Conditional model:
## Estimate Std. Error
## (Intercept) 120.263 92.893
## Spod_perc_survival -10.646 170.497
## SpeciesArtemisia_californica -85.639 97.918
## SpeciesDiplacus_aurantiacus -112.639 99.682
## SpeciesEncelia_californica -96.425 99.307
## SpeciesEriogonum_fasciculatum -118.100 114.408
## SpeciesGrindelia_camporum -108.051 234.427
## SpeciesIsocoma_menziesii -112.864 107.480
## SpeciesMalacothamnus_fasciculatus -105.080 104.654
## SpeciesMalacothrix_saxatilis -26.360 104.654
## SpeciesMirabilis_laevis 126.652 98.528
## SpeciesSalvia_apiana -101.282 116.473
## SpeciesSalvia_mellifera -98.250 101.616
## SpeciesSisyrinchium_bellum -116.964 108.666
## SpeciesStachys_ajugoides_var_rigida -112.634 103.858
## Spod_perc_survival:SpeciesArtemisia_californica -43.736 427.185
## Spod_perc_survival:SpeciesDiplacus_aurantiacus 22.614 204.010
## Spod_perc_survival:SpeciesEncelia_californica -33.717 375.301
## Spod_perc_survival:SpeciesEriogonum_fasciculatum 51.074 201.753
## Spod_perc_survival:SpeciesGrindelia_camporum NA NA
## Spod_perc_survival:SpeciesIsocoma_menziesii 6.787 205.986
## Spod_perc_survival:SpeciesMalacothamnus_fasciculatus 7.476 246.849
## Spod_perc_survival:SpeciesMalacothrix_saxatilis -89.731 235.838
## Spod_perc_survival:SpeciesMirabilis_laevis -543.301 567.855
## Spod_perc_survival:SpeciesSalvia_apiana -5.862 197.550
## Spod_perc_survival:SpeciesSalvia_mellifera 61.153 201.752
## Spod_perc_survival:SpeciesSisyrinchium_bellum NA NA
## Spod_perc_survival:SpeciesStachys_ajugoides_var_rigida NA NA
## z value Pr(>|z|)
## (Intercept) 1.295 0.195
## Spod_perc_survival -0.062 0.950
## SpeciesArtemisia_californica -0.875 0.382
## SpeciesDiplacus_aurantiacus -1.130 0.258
## SpeciesEncelia_californica -0.971 0.332
## SpeciesEriogonum_fasciculatum -1.032 0.302
## SpeciesGrindelia_camporum -0.461 0.645
## SpeciesIsocoma_menziesii -1.050 0.294
## SpeciesMalacothamnus_fasciculatus -1.004 0.315
## SpeciesMalacothrix_saxatilis -0.252 0.801
## SpeciesMirabilis_laevis 1.285 0.199
## SpeciesSalvia_apiana -0.870 0.385
## SpeciesSalvia_mellifera -0.967 0.334
## SpeciesSisyrinchium_bellum -1.076 0.282
## SpeciesStachys_ajugoides_var_rigida -1.084 0.278
## Spod_perc_survival:SpeciesArtemisia_californica -0.102 0.918
## Spod_perc_survival:SpeciesDiplacus_aurantiacus 0.111 0.912
## Spod_perc_survival:SpeciesEncelia_californica -0.090 0.928
## Spod_perc_survival:SpeciesEriogonum_fasciculatum 0.253 0.800
## Spod_perc_survival:SpeciesGrindelia_camporum NA NA
## Spod_perc_survival:SpeciesIsocoma_menziesii 0.033 0.974
## Spod_perc_survival:SpeciesMalacothamnus_fasciculatus 0.030 0.976
## Spod_perc_survival:SpeciesMalacothrix_saxatilis -0.380 0.704
## Spod_perc_survival:SpeciesMirabilis_laevis -0.957 0.339
## Spod_perc_survival:SpeciesSalvia_apiana -0.030 0.976
## Spod_perc_survival:SpeciesSalvia_mellifera 0.303 0.762
## Spod_perc_survival:SpeciesSisyrinchium_bellum NA NA
## Spod_perc_survival:SpeciesStachys_ajugoides_var_rigida NA NA## look at plot of model
plot_model(herbmass_surv, type = "pred", terms = c("Spod_perc_survival", "Species"), se - T,ci.lvl = .5)+
ggtitle("Predicted herbivore mass as a function of spodoptera survival")## Warning in RColorBrewer::brewer.pal(n, pal): n too large, allowed maximum for palette Set1 is 9
## Returning the palette you asked for with that many colors
TAKEAWAY Herbivore mass shows similar patterns to total mass in that there is no significant difference in mass as a function of spodoptera survival. Since mass is not changing but count is, the mass of each arthropod must be changing.
Let’s look at predators. Earlier, predator counts showed no correlation with spodoptera survival.
2.4.4 predator mass ~ Spodoptera survival
## Run glmmTMB model for predator mass ~ Spodoptera percent survival. Family = gaussian.
predmass_surv <- glmmTMB(Mass_Pred_mg ~ Spod_perc_survival * Species + (1|Plot) + (1|Watering_Treatment) + (1|Plant), data = total_plant, family = gaussian())## dropping columns from rank-deficient conditional model: Spod_perc_survival:SpeciesGrindelia_camporum, Spod_perc_survival:SpeciesSisyrinchium_bellum, Spod_perc_survival:SpeciesStachys_ajugoides_var_rigida## Warning in finalizeTMB(TMBStruc, obj, fit, h, data.tmb.old): Model convergence
## problem; non-positive-definite Hessian matrix. See vignette('troubleshooting')##Test dispersion with Dharma package.
testDispersion(predmass_surv)
##
## DHARMa nonparametric dispersion test via sd of residuals fitted vs.
## simulated
##
## data: simulationOutput
## dispersion = 1.0032, p-value = 0.944
## alternative hypothesis: two.sided## results indicate the data is not over- or under-dispersed and we can stick with gaussian
Anova(predmass_surv, type = 2)| Chisq | Df | Pr(>Chisq) |
|---|---|---|
| 0.105 | 1 | 0.746 |
| 34.1 | 13 | 0.00117 |
| 16.8 | 10 | 0.079 |
summary(predmass_surv)## Family: gaussian ( identity )
## Formula:
## Mass_Pred_mg ~ Spod_perc_survival * Species + (1 | Plot) + (1 |
## Watering_Treatment) + (1 | Plant)
## Data: total_plant
##
## AIC BIC logLik deviance df.resid
## NA NA NA NA 141
##
## Random effects:
##
## Conditional model:
## Groups Name Variance Std.Dev.
## Plot (Intercept) 2.661e-01 0.51589
## Watering_Treatment (Intercept) 2.757e-05 0.00525
## Plant (Intercept) 3.433e+01 5.85878
## Residual 2.797e-01 0.52889
## Number of obs: 170, groups: Plot, 20; Watering_Treatment, 6; Plant, 170
##
## Dispersion estimate for gaussian family (sigma^2): 0.28
##
## Conditional model:
## Estimate Std. Error
## (Intercept) 2.12306 4.24906
## Spod_perc_survival 0.17627 7.69957
## SpeciesArtemisia_californica 1.46273 4.47059
## SpeciesDiplacus_aurantiacus 1.47138 4.57849
## SpeciesEncelia_californica 2.63132 4.52818
## SpeciesEriogonum_fasciculatum -0.11499 5.17905
## SpeciesGrindelia_camporum -1.63612 10.58414
## SpeciesIsocoma_menziesii 13.25173 4.89272
## SpeciesMalacothamnus_fasciculatus 0.02283 4.80808
## SpeciesMalacothrix_saxatilis 1.26479 4.77049
## SpeciesMirabilis_laevis -1.10010 4.52236
## SpeciesSalvia_apiana -1.24288 5.32563
## SpeciesSalvia_mellifera 2.05686 4.64782
## SpeciesSisyrinchium_bellum -1.69602 4.90617
## SpeciesStachys_ajugoides_var_rigida -1.67871 4.73647
## Spod_perc_survival:SpeciesArtemisia_californica 13.54421 19.17142
## Spod_perc_survival:SpeciesDiplacus_aurantiacus -3.98756 9.17741
## Spod_perc_survival:SpeciesEncelia_californica -13.61942 16.87919
## Spod_perc_survival:SpeciesEriogonum_fasciculatum -1.14643 9.13031
## Spod_perc_survival:SpeciesGrindelia_camporum NA NA
## Spod_perc_survival:SpeciesIsocoma_menziesii -14.25247 9.28778
## Spod_perc_survival:SpeciesMalacothamnus_fasciculatus -1.30333 11.10473
## Spod_perc_survival:SpeciesMalacothrix_saxatilis -5.33828 10.65355
## Spod_perc_survival:SpeciesMirabilis_laevis 5.49813 25.59626
## Spod_perc_survival:SpeciesSalvia_apiana 2.45431 8.88709
## Spod_perc_survival:SpeciesSalvia_mellifera 12.36784 9.07727
## Spod_perc_survival:SpeciesSisyrinchium_bellum NA NA
## Spod_perc_survival:SpeciesStachys_ajugoides_var_rigida NA NA
## z value Pr(>|z|)
## (Intercept) 0.500 0.61732
## Spod_perc_survival 0.023 0.98174
## SpeciesArtemisia_californica 0.327 0.74352
## SpeciesDiplacus_aurantiacus 0.321 0.74793
## SpeciesEncelia_californica 0.581 0.56117
## SpeciesEriogonum_fasciculatum -0.022 0.98229
## SpeciesGrindelia_camporum -0.155 0.87715
## SpeciesIsocoma_menziesii 2.708 0.00676 **
## SpeciesMalacothamnus_fasciculatus 0.005 0.99621
## SpeciesMalacothrix_saxatilis 0.265 0.79091
## SpeciesMirabilis_laevis -0.243 0.80781
## SpeciesSalvia_apiana -0.233 0.81547
## SpeciesSalvia_mellifera 0.442 0.65810
## SpeciesSisyrinchium_bellum -0.346 0.72957
## SpeciesStachys_ajugoides_var_rigida -0.354 0.72302
## Spod_perc_survival:SpeciesArtemisia_californica 0.707 0.47989
## Spod_perc_survival:SpeciesDiplacus_aurantiacus -0.434 0.66393
## Spod_perc_survival:SpeciesEncelia_californica -0.807 0.41974
## Spod_perc_survival:SpeciesEriogonum_fasciculatum -0.126 0.90008
## Spod_perc_survival:SpeciesGrindelia_camporum NA NA
## Spod_perc_survival:SpeciesIsocoma_menziesii -1.534 0.12490
## Spod_perc_survival:SpeciesMalacothamnus_fasciculatus -0.117 0.90657
## Spod_perc_survival:SpeciesMalacothrix_saxatilis -0.501 0.61631
## Spod_perc_survival:SpeciesMirabilis_laevis 0.215 0.82992
## Spod_perc_survival:SpeciesSalvia_apiana 0.276 0.78242
## Spod_perc_survival:SpeciesSalvia_mellifera 1.362 0.17304
## Spod_perc_survival:SpeciesSisyrinchium_bellum NA NA
## Spod_perc_survival:SpeciesStachys_ajugoides_var_rigida NA NA
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1## look at plot of model
plot_model(predmass_surv, type = "pred", terms = c("Spod_perc_survival", "Species"), se - T,ci.lvl = .5)+
ggtitle("Predicted predator mass as a function of spodoptera survival")## Warning in RColorBrewer::brewer.pal(n, pal): n too large, allowed maximum for palette Set1 is 9
## Returning the palette you asked for with that many colors
TAKEAWAY There is no signifant relationship between spodoptera survival and predator mass with only 1 species showing significance (Isocoma). Predators seem to be truly independent of anything related to spodoptera survival.
New side-quest: convert dataset from wide to long, create new columns title “order” and “biomass” so that we can incorporate order as a random effect in the model.
## Model tinkering
### Consider dropping plot and watering treatment as random effects as they have little overall effect on the model. KEEP plant, though.
### Consider dropping ACMGLA and STAAJU from the dataset
### Remove the interactive term of yearKailen’s to-do list
- Remove log-transformation from avg. spodoptera mass as it is not necessary that dependent variables are normally distributed, rerun lmer model. Consider modifying model to arth ~ spod mass + year + spod * year (I DONT UNDERSTAND WHY SPECIES NOT INCLUDED)
- Figure out how to display ANOVA results in a better table that lists individual species effects on measured variable.
- Plot results of linear regression model according to species for both spodoptera mass and survival, assuming that species-level effects are significant. Use raw data, include r^2 and p-values (14 separate plots). Fix axes scale to be the same across all species.
- Remove dead spodoptera mass from survival results (remove 0’s and convert to ‘na’)
- Get rid of tab_model, effect and whisker plots as they are not necessary.
- Figure out way to rerun survival model with each caterpillar as an individual as a data point as opposed to percent survival.
Nell, Colleen S., and Kailen A. Mooney. 2019. “Plant Structural
Complexity Mediates Trade-Off in Direct and Indirect Plant
Defense by Birds.” Ecology 100 (10). https://doi.org/10.1002/ecy.2853.
Pratt, Jessica D., Andrew Datu, Thi Tran, Daniel C. Sheng, and Kailen A.
Mooney. 2017. “Genetically Based Latitudinal Clines in
Artemisia Californica Drive Parallel Clines in
Arthropod Communities.” Ecology 98 (1): 79–91. https://doi.org/10.1002/ecy.1620.
Rogers, Lee E., W. T. Hinds, and Ray L. Buschbom. 1976. “A General
Weight Vs. Length Relationship for Insects1.” Annals of the
Entomological Society of America 69 (2): 387–89. https://doi.org/10.1093/aesa/69.2.387.
You, Yanrong, Chunpeng An, and Chuanyou Li. 2020. “Insect Feeding
Assays with Spodoptera Exigua on Arabidopsis Thaliana.”
BIO-PROTOCOL 10 (5). https://doi.org/10.21769/BioProtoc.3538.