“A controlled microclimate contrast of three native and one highly invasive plant species in Southern California”

A controlled microclimate contrast of three native and one highly invasive plant species in Southern California Mario Zuliani1*, Laura Brussa, Jessica Cunsolo, Angela Zuliani, and Christopher J. Lortie1. 1Department of Biological Science, York University, 4700 Keele St, Toronto, ON M3J 1P3, Canada Question: How do increasing ambient temperature in arid ecosystems influence both the successive germination and total biomass of 3 natives and 1 exotic annual plant species.

Location: Greenhouse simulation conditions within the Carrizo Plain National Monument, California, USA (35.117985, -119.608762)

Methods: The effects of temperature were tested on 3 native, Salvia columbariae, Layia platyglossa, Phacelia tanacetifolia and 1 exotic plant species, Bromus rubens independently over a 6-week growing period in a temperature controlled greenhouse in 2021. 210 individual replicates were conducted across an increasing temperature gradient meant to simulate the arid ecosystems of Southern California. Hourly temperature was recorded through ambient temperature pendants. Annual biomass was then recorded at the conclusion of each 6-week trial.

Results: The evaluation of temperature on the overall successive germination of native and exotic annual plant species showed an overall negative affect. Increasing temperatures negatively influenced the total germination of one native – Layia platyglossa - and one exotic – Bromus rubens - annual species. Regarding individual biomass, increases in ambient temperature negatively influenced both the overall size and biomass of 2 native – Layia platyglossa & Salvia columbaria – and one exotic – Bromus rubens - annual species.

Conclusion: These findings suggest that increases in ambient temperature due to global warming can have negative impacts on both native and exotic plant establishment and succession. Analyzing the performance and establishment of these annual species is essential to understanding local plant community composition while determining the responses both native and exotic annuals have to increasing abiotic stressors within arid/semi-arid ecosystems. Specifically understanding how increasing ambient temperature influences exotic annual establishment can be vital knowledge for managing the spread and establishment of these species.

Key Words: Intraspecific Association, Desert, Grassland, Greenhouse, Temperature, Bromus rubens, Exotic Species, Restoration

###Packages for future work
library(rmarkdown)
library(ggmap)
## Loading required package: ggplot2
## ℹ Google's Terms of Service: <https://mapsplatform.google.com>
## ℹ Please cite ggmap if you use it! Use `citation("ggmap")` for details.
library(ggplot2)
library(tidyverse)
## ── Attaching packages
## ───────────────────────────────────────
## tidyverse 1.3.2 ──
## ✔ tibble  3.1.8      ✔ dplyr   1.0.10
## ✔ tidyr   1.2.1      ✔ stringr 1.4.1 
## ✔ readr   2.1.3      ✔ forcats 0.5.2 
## ✔ purrr   0.3.5      
## ── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
## ✖ dplyr::filter() masks stats::filter()
## ✖ dplyr::lag()    masks stats::lag()
library(dplyr)
library(MASS)
## 
## Attaching package: 'MASS'
## 
## The following object is masked from 'package:dplyr':
## 
##     select
library(ggpubr) 
library(emmeans)
##Prep Temp Data
#Layia <- read.csv("All Temp Data.csv")
#Brome <- read.csv("Brome Temp Sheet.csv")
#Salvia <- read.csv("Salvia Temp Sheet.csv")
#Phacelia <- read.csv("Phacelia Temp Sheet.csv")

###Merge all species temp data
#Temp <- merge(Layia, Brome, all = TRUE)
#Temp <- merge(Temp, Salvia, all = TRUE)
#Temp <- merge(Temp, Phacelia, all = TRUE)

###Set up SD and SE
#se_temp <- sd(Temp$temperature)/sqrt(length(Temp$temperature))
#Temp$se <- se_temp

###Get mean and max temp
#Temp <- Temp %>%
  #group_by(as.character(temp), species, pendant_ID, se) %>%
  #summarise(mean_temp = mean(temperature), max_temp = max(temperature))
#names(Temp)[1] <- "temp"

###Output file
#write.csv(Temp, "Temp.csv")
###Clean up Temperature and determine the mean and max for each pendant.
#Temp2 <- read.csv("Temp_2.csv")
#Prep Germination Data and Combine With Temp Data
#Germ <- read.csv("Final Germination.csv")
#final <- merge(Germ, Temp2, all = TRUE)

#write.csv(final, "final.csv")
final <- read.csv("final.csv")

###Data Viz

###Temperature vs Number of Germinated Individuals in 6 weeks by species
TempFactor <- ggplot(final, aes(temp, germination),show.legend=FALSE) +
  geom_boxplot() +
  facet_wrap(~species)+
  scale_color_brewer(palette = "Set1") + theme_classic() + labs(tag = "")+
  theme(axis.title.x = element_blank()) +
  geom_smooth(method = lm, se = TRUE) + 
  labs(x = "Temperature", y = "Number of Germinated Seeds")
TempFactor
## `geom_smooth()` using formula = 'y ~ x'

###Plotting the lm
plot(lm(germination ~ mean_temp, data = final), which = 1)

germination_poly <- lm(germination ~ mean_temp + I(mean_temp^2), data = final)

library(car)
## Loading required package: carData
## 
## Attaching package: 'car'
## The following object is masked from 'package:dplyr':
## 
##     recode
## The following object is masked from 'package:purrr':
## 
##     some
library(broom)
Anova(germination_poly)
## Anova Table (Type II tests)
## 
## Response: germination
##                Sum Sq  Df F value   Pr(>F)   
## mean_temp         946   1  7.6086 0.005936 **
## I(mean_temp^2)    916   1  7.3724 0.006760 **
## Residuals      104027 837                    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
###General Figure showing Temp vs total germination in 6 weeks (Not useable)
Temp <- ggplot(final, aes(mean_temp, germination),show.legend=FALSE) +
  geom_point() +
  scale_color_brewer(palette = "Set1") + theme_classic() + 
  geom_smooth(method = lm, se = TRUE) + 
  labs(x = "Temperature", y = "Number of Germinated Seeds")
Temp
## `geom_smooth()` using formula = 'y ~ x'

###Figure showing Temp vs germination by species (Not facetted)
Tempspecies <- ggplot(final, aes(mean_temp, germination, color = species),show.legend=FALSE) +
  geom_point() +
  scale_color_brewer(palette = "Set1") + theme_classic() + 
  geom_smooth(method = lm, se = TRUE) + 
  labs(x = "Temperature", y = "Number of Germinated Seeds")
Tempspecies
## `geom_smooth()` using formula = 'y ~ x'

###This is Temperature vs germination, facet by species. I am not the biggest fan of this figure though
TempspeciesFacet <- ggplot(final, aes(mean_temp, germination, color = species),show.legend=FALSE) +
  geom_point() +
  facet_wrap(~temp)+
  scale_color_brewer(palette = "Set1") + theme_classic() + 
  geom_smooth(method = lm, se = TRUE) + 
  labs(x = "Temperature", y = "Number of Germinated Seeds")
TempspeciesFacet <- TempspeciesFacet + theme(legend.title = element_text(size = 3), 
               legend.text = element_text(size = 3))
TempspeciesFacet
## `geom_smooth()` using formula = 'y ~ x'

ggplot(shelter.shrub.open, aes((day), temp, color=microsite)) + geom_smooth()+ xlab(“Day”) + ylab (“Temperature (°F)”)+ theme_classic()+ theme(axis.text=element_text(size=12))+stat_summary(fun.y=max, geom=“point”, size=2, aes(shape = microsite))+ labs(color=“Microsite”, shape= “Microsite”)

###Plot for Temperature and germination in 6 weeks by species
TempspeciesFacet <- ggplot(final, aes(mean_temp, germination),show.legend=FALSE) +
  geom_smooth(method = lm) + 
  scale_color_brewer(palette = "Set1") + theme_classic() + 
    facet_wrap(~species)+
  labs(x = "Mean Temperature (°C)", y = "Mean Number of Germinations") +stat_summary(fun.y=mean, geom="point", size=1.5)
## Warning: The `fun.y` argument of `stat_summary()` is deprecated as of ggplot2 3.3.0.
## ℹ Please use the `fun` argument instead.
TempspeciesFacet
## `geom_smooth()` using formula = 'y ~ x'

###Figure shows temperature v Germination in 6 weeks, facet by species and filled via temperature
TempspeciesFacetColor <- ggplot(final, aes(mean_temp, germination, color = temp),show.legend=FALSE) +
  geom_point() +
  facet_wrap(~species)+
  scale_color_brewer(palette = "Set1") + theme_classic() +
  geom_smooth(method = lm, se = TRUE) + 
  labs(x = "Temperature", y = "Number of Germinated Seeds")
TempspeciesFacetColor
## `geom_smooth()` using formula = 'y ~ x'

###Statistics For Germination of Each species at varying temperatures
m1 <- glm(germination ~ temp*species+mean_temp, family = "quasipoisson", data = final)
anova(m1, test = "Chisq")
## Analysis of Deviance Table
## 
## Model: quasipoisson, link: log
## 
## Response: germination
## 
## Terms added sequentially (first to last)
## 
## 
##              Df Deviance Resid. Df Resid. Dev  Pr(>Chi)    
## NULL                           839    10186.2              
## temp          2    275.3       837     9910.9 < 2.2e-16 ***
## species       3   7283.5       834     2627.4 < 2.2e-16 ***
## mean_temp     1     10.1       833     2617.3   0.06068 .  
## temp:species  6    208.3       827     2409.0 1.321e-13 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
###Was temp manipulated significantly in the experiment?
e1 <- emmeans(m1, pairwise~temp)
## NOTE: Results may be misleading due to involvement in interactions
e1
## $emmeans
##  temp   emmean     SE  df asymp.LCL asymp.UCL
##  High     1.62 0.0824 Inf      1.46      1.78
##  Low      1.97 0.0614 Inf      1.85      2.09
##  Medium   1.83 0.0590 Inf      1.71      1.94
## 
## Results are averaged over the levels of: species 
## Results are given on the log (not the response) scale. 
## Confidence level used: 0.95 
## 
## $contrasts
##  contrast      estimate     SE  df z.ratio p.value
##  High - Low      -0.351 0.1164 Inf  -3.015  0.0072
##  High - Medium   -0.206 0.1113 Inf  -1.851  0.1533
##  Low - Medium     0.145 0.0771 Inf   1.879  0.1449
## 
## Results are averaged over the levels of: species 
## Results are given on the log (not the response) scale. 
## P value adjustment: tukey method for comparing a family of 3 estimates
###Germination by temp and species
e2 <- emmeans(m1, pairwise~temp|species) ###Low Temp brome germinates the same amount as Layia at all temps
e2
## $emmeans
## species = Bromus rubens:
##  temp    emmean     SE  df asymp.LCL asymp.UCL
##  High    1.2080 0.1355 Inf     0.943     1.474
##  Low     2.1947 0.0691 Inf     2.059     2.330
##  Medium  1.5953 0.0908 Inf     1.417     1.773
## 
## species = Layia platyglossa:
##  temp    emmean     SE  df asymp.LCL asymp.UCL
##  High    1.9438 0.0836 Inf     1.780     2.108
##  Low     2.3695 0.0736 Inf     2.225     2.514
##  Medium  2.2819 0.0788 Inf     2.127     2.436
## 
## species = Phacelia tanacetifolia:
##  temp    emmean     SE  df asymp.LCL asymp.UCL
##  High    3.3623 0.0817 Inf     3.202     3.522
##  Low     3.2934 0.0449 Inf     3.205     3.381
##  Medium  3.3526 0.0379 Inf     3.278     3.427
## 
## species = Salvia columbariae:
##  temp    emmean     SE  df asymp.LCL asymp.UCL
##  High   -0.0246 0.2208 Inf    -0.457     0.408
##  Low     0.0355 0.1933 Inf    -0.343     0.414
##  Medium  0.0837 0.1892 Inf    -0.287     0.455
## 
## Results are given on the log (not the response) scale. 
## Confidence level used: 0.95 
## 
## $contrasts
## species = Bromus rubens:
##  contrast      estimate     SE  df z.ratio p.value
##  High - Low    -0.98667 0.1604 Inf  -6.152  <.0001
##  High - Medium -0.38724 0.1650 Inf  -2.347  0.0496
##  Low - Medium   0.59943 0.1131 Inf   5.298  <.0001
## 
## species = Layia platyglossa:
##  contrast      estimate     SE  df z.ratio p.value
##  High - Low    -0.42574 0.1214 Inf  -3.508  0.0013
##  High - Medium -0.33812 0.1257 Inf  -2.690  0.0195
##  Low - Medium   0.08762 0.0841 Inf   1.042  0.5501
## 
## species = Phacelia tanacetifolia:
##  contrast      estimate     SE  df z.ratio p.value
##  High - Low     0.06895 0.1100 Inf   0.627  0.8054
##  High - Medium  0.00973 0.0925 Inf   0.105  0.9939
##  Low - Medium  -0.05922 0.0574 Inf  -1.031  0.5572
## 
## species = Salvia columbariae:
##  contrast      estimate     SE  df z.ratio p.value
##  High - Low    -0.06010 0.2976 Inf  -0.202  0.9778
##  High - Medium -0.10830 0.2945 Inf  -0.368  0.9282
##  Low - Medium  -0.04819 0.2674 Inf  -0.180  0.9823
## 
## Results are given on the log (not the response) scale. 
## P value adjustment: tukey method for comparing a family of 3 estimates
###Difference in germination by species and temp
e3 <- emmeans(m1, pairwise~species|temp) ###Low Temp brome germinates the same amount as Layia at all temps
e3
## $emmeans
## temp = High:
##  species                 emmean     SE  df asymp.LCL asymp.UCL
##  Bromus rubens           1.2080 0.1355 Inf     0.943     1.474
##  Layia platyglossa       1.9438 0.0836 Inf     1.780     2.108
##  Phacelia tanacetifolia  3.3623 0.0817 Inf     3.202     3.522
##  Salvia columbariae     -0.0246 0.2208 Inf    -0.457     0.408
## 
## temp = Low:
##  species                 emmean     SE  df asymp.LCL asymp.UCL
##  Bromus rubens           2.1947 0.0691 Inf     2.059     2.330
##  Layia platyglossa       2.3695 0.0736 Inf     2.225     2.514
##  Phacelia tanacetifolia  3.2934 0.0449 Inf     3.205     3.381
##  Salvia columbariae      0.0355 0.1933 Inf    -0.343     0.414
## 
## temp = Medium:
##  species                 emmean     SE  df asymp.LCL asymp.UCL
##  Bromus rubens           1.5953 0.0908 Inf     1.417     1.773
##  Layia platyglossa       2.2819 0.0788 Inf     2.127     2.436
##  Phacelia tanacetifolia  3.3526 0.0379 Inf     3.278     3.427
##  Salvia columbariae      0.0837 0.1892 Inf    -0.287     0.455
## 
## Results are given on the log (not the response) scale. 
## Confidence level used: 0.95 
## 
## $contrasts
## temp = High:
##  contrast                                    estimate     SE  df z.ratio
##  Bromus rubens - Layia platyglossa             -0.736 0.1490 Inf  -4.937
##  Bromus rubens - Phacelia tanacetifolia        -2.154 0.1281 Inf -16.811
##  Bromus rubens - Salvia columbariae             1.233 0.2494 Inf   4.942
##  Layia platyglossa - Phacelia tanacetifolia    -1.419 0.1003 Inf -14.138
##  Layia platyglossa - Salvia columbariae         1.968 0.2317 Inf   8.494
##  Phacelia tanacetifolia - Salvia columbariae    3.387 0.2232 Inf  15.174
##  p.value
##   <.0001
##   <.0001
##   <.0001
##   <.0001
##   <.0001
##   <.0001
## 
## temp = Low:
##  contrast                                    estimate     SE  df z.ratio
##  Bromus rubens - Layia platyglossa             -0.175 0.0910 Inf  -1.921
##  Bromus rubens - Phacelia tanacetifolia        -1.099 0.0759 Inf -14.481
##  Bromus rubens - Salvia columbariae             2.159 0.2021 Inf  10.686
##  Layia platyglossa - Phacelia tanacetifolia    -0.924 0.0724 Inf -12.753
##  Layia platyglossa - Salvia columbariae         2.334 0.2000 Inf  11.668
##  Phacelia tanacetifolia - Salvia columbariae    3.258 0.1946 Inf  16.738
##  p.value
##   0.2191
##   <.0001
##   <.0001
##   <.0001
##   <.0001
##   <.0001
## 
## temp = Medium:
##  contrast                                    estimate     SE  df z.ratio
##  Bromus rubens - Layia platyglossa             -0.687 0.1181 Inf  -5.815
##  Bromus rubens - Phacelia tanacetifolia        -1.757 0.0982 Inf -17.889
##  Bromus rubens - Salvia columbariae             1.512 0.2092 Inf   7.224
##  Layia platyglossa - Phacelia tanacetifolia    -1.071 0.0856 Inf -12.511
##  Layia platyglossa - Salvia columbariae         2.198 0.1982 Inf  11.089
##  Phacelia tanacetifolia - Salvia columbariae    3.269 0.1926 Inf  16.977
##  p.value
##   <.0001
##   <.0001
##   <.0001
##   <.0001
##   <.0001
##   <.0001
## 
## Results are given on the log (not the response) scale. 
## P value adjustment: tukey method for comparing a family of 4 estimates
###Mass by Temperature 
Mass <- ggplot(final, aes(mean_temp, mass),show.legend=FALSE) +
  geom_point() +
  scale_color_brewer(palette = "Set1") + theme_classic() + 
  theme() +
  geom_smooth(method = lm, se = TRUE) + 
  labs(x = "Mean Temperature", y = "Mass")
Mass
## `geom_smooth()` using formula = 'y ~ x'

MassSpecies <- ggplot(final, aes(mean_temp, mass, color = species),show.legend=FALSE) +
   geom_smooth(method = lm, se = TRUE) +
  scale_color_brewer(palette = "Set1") + theme_classic() + 
  stat_summary(fun.y=mean, geom="point", size=2)+
  labs(x = "Mean Temperature (°C)", y = "Mean Mass (g)")
MassSpecies
## `geom_smooth()` using formula = 'y ~ x'

MassSpeciesFacet <- ggplot(final, aes(mean_temp, mass),show.legend=FALSE) +
  geom_point() +
  facet_wrap(~species)+
  scale_color_brewer(palette = "Set1") + theme_classic() +
  theme(axis.title.x = element_blank()) +
  geom_smooth(method = lm, se = TRUE) + 
  labs(x = "Temperature", y = "Mass (g)")
MassSpeciesFacet
## `geom_smooth()` using formula = 'y ~ x'

###Temperature v biomass, facet by species and fill by Temp
MassSpeciesFacetColor <- ggplot(final, aes(mean_temp, mass, color = temp),show.legend=FALSE) +
  geom_point() +
  facet_wrap(~species)+
  scale_color_brewer(palette = "Set1") + theme_classic() +
  theme(axis.title.x = element_blank()) +
  geom_smooth(method = lm, se = TRUE) + 
  labs(x = "Temperature", y = "Biomass (g)")
MassSpeciesFacetColor
## `geom_smooth()` using formula = 'y ~ x'

#Use guasian
###Biomass by temp and species
m2 <- glm(mass ~ germination*temp*species+mean_temp, family = "quasipoisson", data = final)
anova(m2, test = "Chisq")
## Analysis of Deviance Table
## 
## Model: quasipoisson, link: log
## 
## Response: mass
## 
## Terms added sequentially (first to last)
## 
## 
##                          Df Deviance Resid. Df Resid. Dev  Pr(>Chi)    
## NULL                                       839    309.227              
## germination               1  175.212       838    134.016 < 2.2e-16 ***
## temp                      2    9.921       836    124.094 < 2.2e-16 ***
## species                   3   16.556       833    107.538 < 2.2e-16 ***
## mean_temp                 1    0.499       832    107.039   0.03672 *  
## germination:temp          2    3.034       830    104.005 1.731e-06 ***
## germination:species       3   13.994       827     90.011 < 2.2e-16 ***
## temp:species              6    6.269       821     83.742 5.050e-10 ***
## germination:temp:species  6    0.583       815     83.159   0.53167    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
e4 <- emmeans(m2, pairwise~temp|species)
## NOTE: Results may be misleading due to involvement in interactions
e4
## $emmeans
## species = Bromus rubens:
##  temp   emmean    SE  df asymp.LCL asymp.UCL
##  High   -2.604 0.294 Inf    -3.181    -2.027
##  Low    -2.170 0.125 Inf    -2.415    -1.926
##  Medium -1.913 0.158 Inf    -2.223    -1.603
## 
## species = Layia platyglossa:
##  temp   emmean    SE  df asymp.LCL asymp.UCL
##  High   -2.063 0.135 Inf    -2.328    -1.799
##  Low    -3.619 0.267 Inf    -4.141    -3.096
##  Medium -2.510 0.161 Inf    -2.825    -2.195
## 
## species = Phacelia tanacetifolia:
##  temp   emmean    SE  df asymp.LCL asymp.UCL
##  High   -0.530 0.132 Inf    -0.789    -0.270
##  Low    -0.391 0.285 Inf    -0.949     0.167
##  Medium -0.279 0.229 Inf    -0.727     0.170
## 
## species = Salvia columbariae:
##  temp   emmean    SE  df asymp.LCL asymp.UCL
##  High    1.378 0.660 Inf     0.085     2.672
##  Low    -1.155 0.771 Inf    -2.666     0.356
##  Medium  1.643 0.954 Inf    -0.226     3.512
## 
## Results are given on the log (not the response) scale. 
## Confidence level used: 0.95 
## 
## $contrasts
## species = Bromus rubens:
##  contrast      estimate    SE  df z.ratio p.value
##  High - Low      -0.434 0.325 Inf  -1.336  0.3753
##  High - Medium   -0.691 0.336 Inf  -2.058  0.0988
##  Low - Medium    -0.257 0.200 Inf  -1.284  0.4043
## 
## species = Layia platyglossa:
##  contrast      estimate    SE  df z.ratio p.value
##  High - Low       1.555 0.305 Inf   5.103  <.0001
##  High - Medium    0.447 0.219 Inf   2.040  0.1029
##  Low - Medium    -1.108 0.299 Inf  -3.702  0.0006
## 
## species = Phacelia tanacetifolia:
##  contrast      estimate    SE  df z.ratio p.value
##  High - Low      -0.138 0.319 Inf  -0.433  0.9016
##  High - Medium   -0.251 0.266 Inf  -0.944  0.6126
##  Low - Medium    -0.113 0.365 Inf  -0.309  0.9488
## 
## species = Salvia columbariae:
##  contrast      estimate    SE  df z.ratio p.value
##  High - Low       2.534 1.020 Inf   2.484  0.0347
##  High - Medium   -0.265 1.165 Inf  -0.227  0.9719
##  Low - Medium    -2.798 1.224 Inf  -2.287  0.0576
## 
## Results are given on the log (not the response) scale. 
## P value adjustment: tukey method for comparing a family of 3 estimates
###Difference in biomass by species and temp
e5 <- emmeans(m2, pairwise~species|temp)
## NOTE: Results may be misleading due to involvement in interactions
e5
## $emmeans
## temp = High:
##  species                emmean    SE  df asymp.LCL asymp.UCL
##  Bromus rubens          -2.604 0.294 Inf    -3.181    -2.027
##  Layia platyglossa      -2.063 0.135 Inf    -2.328    -1.799
##  Phacelia tanacetifolia -0.530 0.132 Inf    -0.789    -0.270
##  Salvia columbariae      1.378 0.660 Inf     0.085     2.672
## 
## temp = Low:
##  species                emmean    SE  df asymp.LCL asymp.UCL
##  Bromus rubens          -2.170 0.125 Inf    -2.415    -1.926
##  Layia platyglossa      -3.619 0.267 Inf    -4.141    -3.096
##  Phacelia tanacetifolia -0.391 0.285 Inf    -0.949     0.167
##  Salvia columbariae     -1.155 0.771 Inf    -2.666     0.356
## 
## temp = Medium:
##  species                emmean    SE  df asymp.LCL asymp.UCL
##  Bromus rubens          -1.913 0.158 Inf    -2.223    -1.603
##  Layia platyglossa      -2.510 0.161 Inf    -2.825    -2.195
##  Phacelia tanacetifolia -0.279 0.229 Inf    -0.727     0.170
##  Salvia columbariae      1.643 0.954 Inf    -0.226     3.512
## 
## Results are given on the log (not the response) scale. 
## Confidence level used: 0.95 
## 
## $contrasts
## temp = High:
##  contrast                                    estimate    SE  df z.ratio p.value
##  Bromus rubens - Layia platyglossa             -0.540 0.319 Inf  -1.697  0.3254
##  Bromus rubens - Phacelia tanacetifolia        -2.074 0.308 Inf  -6.732  <.0001
##  Bromus rubens - Salvia columbariae            -3.982 0.715 Inf  -5.568  <.0001
##  Layia platyglossa - Phacelia tanacetifolia    -1.534 0.176 Inf  -8.730  <.0001
##  Layia platyglossa - Salvia columbariae        -3.442 0.669 Inf  -5.142  <.0001
##  Phacelia tanacetifolia - Salvia columbariae   -1.908 0.662 Inf  -2.884  0.0205
## 
## temp = Low:
##  contrast                                    estimate    SE  df z.ratio p.value
##  Bromus rubens - Layia platyglossa              1.448 0.289 Inf   5.008  <.0001
##  Bromus rubens - Phacelia tanacetifolia        -1.779 0.309 Inf  -5.756  <.0001
##  Bromus rubens - Salvia columbariae            -1.015 0.779 Inf  -1.303  0.5611
##  Layia platyglossa - Phacelia tanacetifolia    -3.227 0.387 Inf  -8.334  <.0001
##  Layia platyglossa - Salvia columbariae        -2.463 0.812 Inf  -3.034  0.0129
##  Phacelia tanacetifolia - Salvia columbariae    0.764 0.820 Inf   0.931  0.7881
## 
## temp = Medium:
##  contrast                                    estimate    SE  df z.ratio p.value
##  Bromus rubens - Layia platyglossa              0.597 0.223 Inf   2.676  0.0374
##  Bromus rubens - Phacelia tanacetifolia        -1.634 0.278 Inf  -5.874  <.0001
##  Bromus rubens - Salvia columbariae            -3.556 0.966 Inf  -3.680  0.0013
##  Layia platyglossa - Phacelia tanacetifolia    -2.232 0.279 Inf  -8.008  <.0001
##  Layia platyglossa - Salvia columbariae        -4.153 0.963 Inf  -4.312  0.0001
##  Phacelia tanacetifolia - Salvia columbariae   -1.922 0.981 Inf  -1.960  0.2034
## 
## Results are given on the log (not the response) scale. 
## P value adjustment: tukey method for comparing a family of 4 estimates
#Natives Vs Invasives Germination
#Not useful
#locality <- ggplot(final, aes(mean_temp, germination, color = temp),show.legend=FALSE) +
 # geom_point() +
  #facet_wrap(~locality) +
  #scale_color_brewer(palette = "Set1") + theme_classic() +
  #geom_smooth(method = lm, se = TRUE) + 
  #labs(x = "Temperature", y = "Number of Germinated Seeds")
#locality
#Not useful
#locality2 <- ggplot(final, aes(mean_temp, germination),show.legend=FALSE) +
 # geom_point() +
  #facet_wrap(~locality) +
  #scale_color_brewer(palette = "Set1") + theme_classic() +
  #geom_smooth(method = lm, se = TRUE) + 
  #labs(x = "Temperature", y = "Number of Germinated Seeds")
#locality2
#Native vs Invasive Biomass
###Do not use
#locality3 <- ggplot(final, aes(mean_temp, mass, color = temp),show.legend=FALSE) +
  #geom_point() +
 # facet_wrap(~locality) +
 # scale_color_brewer(palette = "Set1") + theme_classic() +
 # geom_smooth(method = lm, se = TRUE) + 
 # labs(x = "Temperature", y = "Bimass (g)")
#locality3
###Do not use
#locality4 <- ggplot(final, aes(mean_temp, mass),show.legend=FALSE) +
 # geom_point() +
  #facet_wrap(~locality) +
  #scale_color_brewer(palette = "Set1") + theme_classic() +
  #geom_smooth(method = lm, se = TRUE) + 
  #labs(x = "Temperature", y = "Biomass (g)")
#locality4
#Natives vs Invasives Stats Mass
#Not useful
#m3 <- glm(mass ~ germination*temp*locality, family = "quasipoisson", data = final)
#anova(m3, test = "Chisq")
#e6 <- emmeans(m3, pairwise~locality|temp)
#e6
#m4 <- glm(germination ~ temp*locality, family = "quasipoisson", data = final)
#anova(m4, test = "Chisq")
#e7 <- emmeans(m4, pairwise~locality|temp)
#e7

###November 24th 2022 (After meeting)

###Boxplots Comparing Mean and Max Temp Blocks
ggplot(final, aes(x=factor (temp, level=c("Low", "Medium", "High")), mean_temp)) +
  geom_boxplot() + theme_classic() + labs(x = "Temperature Treatment", y = "Mean Temperature")

ggplot(final, aes(x=factor (temp, level=c("Low", "Medium", "High")), max_temp)) +
  geom_boxplot() + theme_classic() + labs(x = "Temperature Treatment", y = "Max Temperature")

###Test Biomass by Established Density with Mean and Max Temp

###Set up Biomass by Established Density
final$average_mass <- final$mass/final$establishment

###By mean Temp
ggplot(final, aes(mean_temp, average_mass),show.legend=FALSE) +
  geom_point() +
  facet_wrap(~species) +
  scale_color_brewer(palette = "Set1") + theme_classic() +
  geom_smooth(method = lm, se = TRUE) + 
  labs(x = "Temperature", y = "Bimass per Individual")
## `geom_smooth()` using formula = 'y ~ x'
## Warning: Removed 265 rows containing non-finite values (`stat_smooth()`).
## Warning: Removed 265 rows containing missing values (`geom_point()`).

###Dont like this one
#ggplot(final, aes(mean_temp, average_mass, color = temp),show.legend=FALSE) +
 # geom_point() +
  #facet_wrap(~species) +
  #scale_color_brewer(palette = "Set1") + theme_classic() +
  #geom_smooth(method = lm, se = TRUE) + 
  #labs(x = "Temperature", y = "Bimass per Individual")

###Max Temp
ggplot(final, aes(max_temp, average_mass),show.legend=FALSE) +
  geom_point() +
  facet_wrap(~species) +
  scale_color_brewer(palette = "Set1") + theme_classic() +
  geom_smooth(method = lm, se = TRUE) + 
  labs(x = "Temperature", y = "Bimass per Individual")
## `geom_smooth()` using formula = 'y ~ x'
## Warning: Removed 265 rows containing non-finite values (`stat_smooth()`).
## Removed 265 rows containing missing values (`geom_point()`).

ggplot(final, aes(max_temp, average_mass, color = temp),show.legend=FALSE) +
  geom_point() +
  facet_wrap(~species) +
  scale_color_brewer(palette = "Set1") + theme_classic() +
  geom_smooth(method = lm, se = TRUE) + 
  labs(x = "Temperature", y = "Bimass per Individual")
## `geom_smooth()` using formula = 'y ~ x'
## Warning: Removed 265 rows containing non-finite values (`stat_smooth()`).
## Removed 265 rows containing missing values (`geom_point()`).

###Ran as Gaussian since temp is blocked
m5 <- glm(average_mass ~ temp*species, family = "gaussian", data = final)
anova(m5, test = "Chisq")
## Analysis of Deviance Table
## 
## Model: gaussian, link: identity
## 
## Response: average_mass
## 
## Terms added sequentially (first to last)
## 
## 
##              Df Deviance Resid. Df Resid. Dev Pr(>Chi)    
## NULL                           574     2.3397             
## temp          2  0.01793       572     2.3218  0.06496 .  
## species       3  0.42287       569     1.8989  < 2e-16 ***
## temp:species  6  0.05304       563     1.8459  0.01283 *  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
e8 <- emmeans(m5, pairwise~temp|species)
e8
## $emmeans
## species = Bromus rubens:
##  temp   emmean      SE  df lower.CL upper.CL
##  High   0.0207 0.00997 563  0.00108   0.0402
##  Low    0.0224 0.00721 563  0.00819   0.0365
##  Medium 0.0370 0.00787 563  0.02154   0.0524
## 
## species = Layia platyglossa:
##  temp   emmean      SE  df lower.CL upper.CL
##  High   0.0342 0.00873 563  0.01710   0.0514
##  Low    0.0659 0.01063 563  0.04500   0.0868
##  Medium 0.0682 0.00772 563  0.05300   0.0833
## 
## species = Phacelia tanacetifolia:
##  temp   emmean      SE  df lower.CL upper.CL
##  High   0.0332 0.00684 563  0.01980   0.0467
##  Low    0.0202 0.00684 563  0.00671   0.0336
##  Medium 0.0209 0.00684 563  0.00745   0.0343
## 
## species = Salvia columbariae:
##  temp   emmean      SE  df lower.CL upper.CL
##  High   0.1191 0.01194 563  0.09566   0.1426
##  Low    0.0864 0.01063 563  0.06548   0.1073
##  Medium 0.1025 0.00941 563  0.08401   0.1210
## 
## Confidence level used: 0.95 
## 
## $contrasts
## species = Bromus rubens:
##  contrast       estimate      SE  df t.ratio p.value
##  High - Low    -0.001706 0.01230 563  -0.139  0.9895
##  High - Medium -0.016329 0.01270 563  -1.286  0.4036
##  Low - Medium  -0.014624 0.01067 563  -1.370  0.3573
## 
## species = Layia platyglossa:
##  contrast       estimate      SE  df t.ratio p.value
##  High - Low    -0.031635 0.01376 563  -2.299  0.0567
##  High - Medium -0.033914 0.01166 563  -2.910  0.0105
##  Low - Medium  -0.002279 0.01314 563  -0.173  0.9836
## 
## species = Phacelia tanacetifolia:
##  contrast       estimate      SE  df t.ratio p.value
##  High - Low     0.013087 0.00968 563   1.352  0.3670
##  High - Medium  0.012349 0.00968 563   1.276  0.4094
##  Low - Medium  -0.000738 0.00968 563  -0.076  0.9968
## 
## species = Salvia columbariae:
##  contrast       estimate      SE  df t.ratio p.value
##  High - Low     0.032746 0.01599 563   2.048  0.1019
##  High - Medium  0.016614 0.01520 563   1.093  0.5190
##  Low - Medium  -0.016132 0.01420 563  -1.136  0.4923
## 
## P value adjustment: tukey method for comparing a family of 3 estimates
m6 <- glm(average_mass ~ mean_temp*species+pendant_ID, family = "quasipoisson", data = final)
anova(m6, test = "Chisq")
## Analysis of Deviance Table
## 
## Model: quasipoisson, link: log
## 
## Response: average_mass
## 
## Terms added sequentially (first to last)
## 
## 
##                   Df Deviance Resid. Df Resid. Dev  Pr(>Chi)    
## NULL                                574     32.104              
## mean_temp          1   0.6515       573     31.452 0.0003796 ***
## species            3   7.8290       570     23.623 < 2.2e-16 ***
## pendant_ID         1   0.6161       569     23.007 0.0005482 ***
## mean_temp:species  3   0.9895       566     22.017 0.0002507 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
e9 <- emmeans(m6, pairwise~species|mean_temp)
e9
## $emmeans
## mean_temp = 26.5:
##  species                emmean     SE  df asymp.LCL asymp.UCL
##  Bromus rubens           -3.59 0.1134 Inf     -3.81     -3.37
##  Layia platyglossa       -3.23 0.1256 Inf     -3.48     -2.99
##  Phacelia tanacetifolia  -3.74 0.1100 Inf     -3.96     -3.53
##  Salvia columbariae      -2.31 0.0784 Inf     -2.46     -2.16
## 
## Results are given on the log (not the response) scale. 
## Confidence level used: 0.95 
## 
## $contrasts
## mean_temp = 26.5:
##  contrast                                    estimate    SE  df z.ratio p.value
##  Bromus rubens - Layia platyglossa             -0.356 0.169 Inf  -2.103  0.1520
##  Bromus rubens - Phacelia tanacetifolia         0.154 0.158 Inf   0.976  0.7634
##  Bromus rubens - Salvia columbariae            -1.279 0.138 Inf  -9.263  <.0001
##  Layia platyglossa - Phacelia tanacetifolia     0.510 0.167 Inf   3.058  0.0120
##  Layia platyglossa - Salvia columbariae        -0.922 0.145 Inf  -6.360  <.0001
##  Phacelia tanacetifolia - Salvia columbariae   -1.433 0.135 Inf -10.606  <.0001
## 
## Results are given on the log (not the response) scale. 
## P value adjustment: tukey method for comparing a family of 4 estimates
###Set up percent successful establishment of individuals
final$succession <- final$establishment/final$germination
###Treat Temp as Block
###Average Mass
ggplot(final, aes(mean_temp, average_mass, color = temp),show.legend=FALSE) +
  geom_point() +
  facet_wrap(~species) +
  scale_color_brewer(palette = "Set1") + theme_classic() +
  geom_smooth(method = lm, se = TRUE) + 
  labs(x = "Temperature", y = "Bimass per Individual")
## `geom_smooth()` using formula = 'y ~ x'
## Warning: Removed 265 rows containing non-finite values (`stat_smooth()`).
## Warning: Removed 265 rows containing missing values (`geom_point()`).

model1 <- glm(average_mass ~ temp*species, family = "quasipoisson", data = final)
anova(model1, test = "Chisq")
## Analysis of Deviance Table
## 
## Model: quasipoisson, link: log
## 
## Response: average_mass
## 
## Terms added sequentially (first to last)
## 
## 
##              Df Deviance Resid. Df Resid. Dev  Pr(>Chi)    
## NULL                           574     32.104              
## temp          2   0.4039       572     31.700  0.022244 *  
## species       3   8.2409       569     23.459 < 2.2e-16 ***
## temp:species  6   1.1449       563     22.314  0.001444 ** 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
em1 <- emmeans(model1, pairwise~temp|species)
em1
## $emmeans
## species = Bromus rubens:
##  temp   emmean    SE  df asymp.LCL asymp.UCL
##  High    -3.88 0.279 Inf     -4.43     -3.33
##  Low     -3.80 0.194 Inf     -4.18     -3.42
##  Medium  -3.30 0.165 Inf     -3.62     -2.97
## 
## species = Layia platyglossa:
##  temp   emmean    SE  df asymp.LCL asymp.UCL
##  High    -3.37 0.190 Inf     -3.75     -3.00
##  Low     -2.72 0.167 Inf     -3.05     -2.39
##  Medium  -2.69 0.119 Inf     -2.92     -2.45
## 
## species = Phacelia tanacetifolia:
##  temp   emmean    SE  df asymp.LCL asymp.UCL
##  High    -3.40 0.151 Inf     -3.70     -3.11
##  Low     -3.90 0.194 Inf     -4.28     -3.52
##  Medium  -3.87 0.190 Inf     -4.24     -3.49
## 
## species = Salvia columbariae:
##  temp   emmean    SE  df asymp.LCL asymp.UCL
##  High    -2.13 0.139 Inf     -2.40     -1.85
##  Low     -2.45 0.146 Inf     -2.73     -2.16
##  Medium  -2.28 0.118 Inf     -2.51     -2.05
## 
## Results are given on the log (not the response) scale. 
## Confidence level used: 0.95 
## 
## $contrasts
## species = Bromus rubens:
##  contrast      estimate    SE  df z.ratio p.value
##  High - Low     -0.0793 0.340 Inf  -0.233  0.9704
##  High - Medium  -0.5825 0.324 Inf  -1.798  0.1701
##  Low - Medium   -0.5032 0.254 Inf  -1.978  0.1177
## 
## species = Layia platyglossa:
##  contrast      estimate    SE  df z.ratio p.value
##  High - Low     -0.6543 0.253 Inf  -2.590  0.0260
##  High - Medium  -0.6883 0.224 Inf  -3.072  0.0060
##  Low - Medium   -0.0340 0.205 Inf  -0.166  0.9849
## 
## species = Phacelia tanacetifolia:
##  contrast      estimate    SE  df z.ratio p.value
##  High - Low      0.5003 0.246 Inf   2.036  0.1037
##  High - Medium   0.4644 0.243 Inf   1.910  0.1357
##  Low - Medium   -0.0360 0.272 Inf  -0.132  0.9904
## 
## species = Salvia columbariae:
##  contrast      estimate    SE  df z.ratio p.value
##  High - Low      0.3215 0.201 Inf   1.596  0.2472
##  High - Medium   0.1502 0.183 Inf   0.822  0.6891
##  Low - Medium   -0.1712 0.188 Inf  -0.913  0.6320
## 
## Results are given on the log (not the response) scale. 
## P value adjustment: tukey method for comparing a family of 3 estimates
###Succession of establishment (Established/germinated)
ggplot(final, aes(mean_temp, succession, color = temp),show.legend=FALSE) +
  geom_point() +
  facet_wrap(~species) +
  scale_color_brewer(palette = "Set1") + theme_classic() +
  geom_smooth(method = lm, se = TRUE) + 
  labs(x = "Temperature", y = "Percent Successfully Established")
## `geom_smooth()` using formula = 'y ~ x'
## Warning: Removed 137 rows containing non-finite values (`stat_smooth()`).
## Warning: Removed 137 rows containing missing values (`geom_point()`).

model2 <- glm(succession ~ temp*species, family = "quasipoisson", data = final)
anova(model2, test = "Chisq")
## Analysis of Deviance Table
## 
## Model: quasipoisson, link: log
## 
## Response: succession
## 
## Terms added sequentially (first to last)
## 
## 
##              Df Deviance Resid. Df Resid. Dev  Pr(>Chi)    
## NULL                           702     383.66              
## temp          2    9.895       700     373.77 6.228e-05 ***
## species       3   43.071       697     330.69 < 2.2e-16 ***
## temp:species  6   58.636       691     272.06 < 2.2e-16 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
em2 <- emmeans(model2, pairwise~temp|species)
em2
## $emmeans
## species = Bromus rubens:
##  temp     emmean     SE  df asymp.LCL asymp.UCL
##  High   -0.55843 0.1458 Inf    -0.844    -0.273
##  Low    -0.31101 0.1020 Inf    -0.511    -0.111
##  Medium -0.45283 0.1138 Inf    -0.676    -0.230
## 
## species = Layia platyglossa:
##  temp     emmean     SE  df asymp.LCL asymp.UCL
##  High    0.01987 0.0899 Inf    -0.156     0.196
##  Low    -2.39802 0.2875 Inf    -2.962    -1.835
##  Medium -1.21443 0.1579 Inf    -1.524    -0.905
## 
## species = Phacelia tanacetifolia:
##  temp     emmean     SE  df asymp.LCL asymp.UCL
##  High   -0.00409 0.0856 Inf    -0.172     0.164
##  Low    -0.00785 0.0858 Inf    -0.176     0.160
##  Medium -0.00230 0.0855 Inf    -0.170     0.165
## 
## species = Salvia columbariae:
##  temp     emmean     SE  df asymp.LCL asymp.UCL
##  High   -0.51935 0.1544 Inf    -0.822    -0.217
##  Low    -0.47163 0.1413 Inf    -0.749    -0.195
##  Medium -0.35771 0.1260 Inf    -0.605    -0.111
## 
## Results are given on the log (not the response) scale. 
## Confidence level used: 0.95 
## 
## $contrasts
## species = Bromus rubens:
##  contrast      estimate    SE  df z.ratio p.value
##  High - Low    -0.24742 0.178 Inf  -1.390  0.3459
##  High - Medium -0.10559 0.185 Inf  -0.571  0.8357
##  Low - Medium   0.14182 0.153 Inf   0.928  0.6227
## 
## species = Layia platyglossa:
##  contrast      estimate    SE  df z.ratio p.value
##  High - Low     2.41789 0.301 Inf   8.027  <.0001
##  High - Medium  1.23429 0.182 Inf   6.793  <.0001
##  Low - Medium  -1.18360 0.328 Inf  -3.608  0.0009
## 
## species = Phacelia tanacetifolia:
##  contrast      estimate    SE  df z.ratio p.value
##  High - Low     0.00375 0.121 Inf   0.031  0.9995
##  High - Medium -0.00179 0.121 Inf  -0.015  0.9999
##  Low - Medium  -0.00554 0.121 Inf  -0.046  0.9988
## 
## species = Salvia columbariae:
##  contrast      estimate    SE  df z.ratio p.value
##  High - Low    -0.04772 0.209 Inf  -0.228  0.9718
##  High - Medium -0.16164 0.199 Inf  -0.811  0.6963
##  Low - Medium  -0.11392 0.189 Inf  -0.602  0.8192
## 
## Results are given on the log (not the response) scale. 
## P value adjustment: tukey method for comparing a family of 3 estimates
###Overall Establishment
ggplot(final, aes(mean_temp, establishment, color = temp),show.legend=FALSE) +
  geom_point() +
  facet_wrap(~species) +
  scale_color_brewer(palette = "Set1") + theme_classic() +
  geom_smooth(method = lm, se = TRUE) + 
  labs(x = "Temperature", y = "Number Established Individuals")
## `geom_smooth()` using formula = 'y ~ x'

model3 <- glm(establishment ~ temp*species, family = "quasipoisson", data = final)
anova(model3, test = "Chisq")
## Analysis of Deviance Table
## 
## Model: quasipoisson, link: log
## 
## Response: establishment
## 
## Terms added sequentially (first to last)
## 
## 
##              Df Deviance Resid. Df Resid. Dev  Pr(>Chi)    
## NULL                           839    12268.9              
## temp          2     71.4       837    12197.6 5.291e-06 ***
## species       3   9552.7       834     2644.9 < 2.2e-16 ***
## temp:species  6    397.5       828     2247.3 < 2.2e-16 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
em3 <- emmeans(model3, pairwise~temp|species)
em3
## $emmeans
## species = Bromus rubens:
##  temp    emmean     SE  df asymp.LCL asymp.UCL
##  High    0.6035 0.1515 Inf     0.307   0.90042
##  Low     2.0263 0.0744 Inf     1.881   2.17203
##  Medium  1.2404 0.1102 Inf     1.025   1.45636
## 
## species = Layia platyglossa:
##  temp    emmean     SE  df asymp.LCL asymp.UCL
##  High    1.3791 0.1028 Inf     1.178   1.58058
##  Low    -0.0144 0.2063 Inf    -0.419   0.38997
##  Medium  1.0194 0.1230 Inf     0.778   1.26051
## 
## species = Phacelia tanacetifolia:
##  temp    emmean     SE  df asymp.LCL asymp.UCL
##  High    3.1588 0.0422 Inf     3.076   3.24156
##  Low     3.3539 0.0383 Inf     3.279   3.42896
##  Medium  3.3594 0.0382 Inf     3.285   3.43423
## 
## species = Salvia columbariae:
##  temp    emmean     SE  df asymp.LCL asymp.UCL
##  High   -0.5108 0.2644 Inf    -1.029   0.00745
##  Low    -0.2412 0.2311 Inf    -0.694   0.21174
##  Medium -0.1710 0.2231 Inf    -0.608   0.26632
## 
## Results are given on the log (not the response) scale. 
## Confidence level used: 0.95 
## 
## $contrasts
## species = Bromus rubens:
##  contrast      estimate     SE  df z.ratio p.value
##  High - Low    -1.42273 0.1687 Inf  -8.431  <.0001
##  High - Medium -0.63691 0.1873 Inf  -3.401  0.0019
##  Low - Medium   0.78582 0.1329 Inf   5.912  <.0001
## 
## species = Layia platyglossa:
##  contrast      estimate     SE  df z.ratio p.value
##  High - Low     1.39351 0.2305 Inf   6.046  <.0001
##  High - Medium  0.35976 0.1603 Inf   2.244  0.0640
##  Low - Medium  -1.03375 0.2402 Inf  -4.304  <.0001
## 
## species = Phacelia tanacetifolia:
##  contrast      estimate     SE  df z.ratio p.value
##  High - Low    -0.19508 0.0570 Inf  -3.423  0.0018
##  High - Medium -0.20056 0.0569 Inf  -3.523  0.0012
##  Low - Medium  -0.00548 0.0541 Inf  -0.101  0.9944
## 
## species = Salvia columbariae:
##  contrast      estimate     SE  df z.ratio p.value
##  High - Low    -0.26966 0.3512 Inf  -0.768  0.7227
##  High - Medium -0.33987 0.3460 Inf  -0.982  0.5880
##  Low - Medium  -0.07020 0.3212 Inf  -0.219  0.9740
## 
## Results are given on the log (not the response) scale. 
## P value adjustment: tukey method for comparing a family of 3 estimates
###Overall Germination

ggplot(final, aes(mean_temp, germination, color = temp),show.legend=FALSE) +
  geom_point() +
  facet_wrap(~species) +
  scale_color_brewer(palette = "Set1") + theme_classic() +
  geom_smooth(method = lm, se = TRUE) + 
  labs(x = "Temperature", y = "Number Germinated Individuals")
## `geom_smooth()` using formula = 'y ~ x'

model4 <- glm(germination ~ temp*species, family = "quasipoisson", data = final)
anova(model4, test = "Chisq")
## Analysis of Deviance Table
## 
## Model: quasipoisson, link: log
## 
## Response: germination
## 
## Terms added sequentially (first to last)
## 
## 
##              Df Deviance Resid. Df Resid. Dev  Pr(>Chi)    
## NULL                           839    10186.2              
## temp          2    275.3       837     9910.9 < 2.2e-16 ***
## species       3   7283.5       834     2627.4 < 2.2e-16 ***
## temp:species  6    195.9       828     2431.5 1.548e-12 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
em4 <- emmeans(model4, pairwise~temp|species)
em4
## $emmeans
## species = Bromus rubens:
##  temp   emmean     SE  df asymp.LCL asymp.UCL
##  High    1.035 0.1217 Inf     0.796     1.273
##  Low     2.253 0.0662 Inf     2.123     2.382
##  Medium  1.609 0.0913 Inf     1.431     1.788
## 
## species = Layia platyglossa:
##  temp   emmean     SE  df asymp.LCL asymp.UCL
##  High    1.870 0.0802 Inf     1.713     2.027
##  Low     2.496 0.0586 Inf     2.381     2.610
##  Medium  2.422 0.0608 Inf     2.303     2.541
## 
## species = Phacelia tanacetifolia:
##  temp   emmean     SE  df asymp.LCL asymp.UCL
##  High    3.162 0.0420 Inf     3.080     3.244
##  Low     3.361 0.0380 Inf     3.286     3.435
##  Medium  3.361 0.0380 Inf     3.287     3.436
## 
## species = Salvia columbariae:
##  temp   emmean     SE  df asymp.LCL asymp.UCL
##  High   -0.138 0.2187 Inf    -0.566     0.291
##  Low     0.121 0.1921 Inf    -0.256     0.498
##  Medium  0.158 0.1886 Inf    -0.211     0.528
## 
## Results are given on the log (not the response) scale. 
## Confidence level used: 0.95 
## 
## $contrasts
## species = Bromus rubens:
##  contrast       estimate     SE  df z.ratio p.value
##  High - Low    -1.218086 0.1385 Inf  -8.795  <.0001
##  High - Medium -0.574729 0.1521 Inf  -3.779  0.0005
##  Low - Medium   0.643356 0.1128 Inf   5.706  <.0001
## 
## species = Layia platyglossa:
##  contrast       estimate     SE  df z.ratio p.value
##  High - Low    -0.625962 0.0993 Inf  -6.304  <.0001
##  High - Medium -0.552669 0.1006 Inf  -5.494  <.0001
##  Low - Medium   0.073293 0.0845 Inf   0.868  0.6606
## 
## species = Phacelia tanacetifolia:
##  contrast       estimate     SE  df z.ratio p.value
##  High - Low    -0.199019 0.0567 Inf  -3.512  0.0013
##  High - Medium -0.199515 0.0567 Inf  -3.522  0.0012
##  Low - Medium  -0.000496 0.0538 Inf  -0.009  1.0000
## 
## species = Salvia columbariae:
##  contrast       estimate     SE  df z.ratio p.value
##  High - Low    -0.258574 0.2911 Inf  -0.888  0.6477
##  High - Medium -0.295845 0.2888 Inf  -1.025  0.5613
##  Low - Medium  -0.037271 0.2692 Inf  -0.138  0.9895
## 
## Results are given on the log (not the response) scale. 
## P value adjustment: tukey method for comparing a family of 3 estimates
###Treating temp as continuous

#average mass by mean_temp
model5 <- glm(average_mass ~ mean_temp*species + pendant_ID, family = "gaussian", data = final)
anova(model5, test = "Chisq")
## Analysis of Deviance Table
## 
## Model: gaussian, link: identity
## 
## Response: average_mass
## 
## Terms added sequentially (first to last)
## 
## 
##                   Df Deviance Resid. Df Resid. Dev  Pr(>Chi)    
## NULL                                574     2.3397              
## mean_temp          1  0.02738       573     2.3124 0.0035771 ** 
## species            3  0.40608       570     1.9063 < 2.2e-16 ***
## pendant_ID         1  0.02722       569     1.8791 0.0036756 ** 
## mean_temp:species  3  0.05311       566     1.8259 0.0009119 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
em5 <- emmeans(model5, pairwise~species|mean_temp)
em5
## $emmeans
## mean_temp = 26.5:
##  species                emmean      SE  df lower.CL upper.CL
##  Bromus rubens          0.0285 0.00469 566   0.0193   0.0377
##  Layia platyglossa      0.0427 0.00591 566   0.0311   0.0543
##  Phacelia tanacetifolia 0.0248 0.00417 566   0.0166   0.0330
##  Salvia columbariae     0.1030 0.00626 566   0.0906   0.1153
## 
## Confidence level used: 0.95 
## 
## $contrasts
## mean_temp = 26.5:
##  contrast                                    estimate      SE  df t.ratio
##  Bromus rubens - Layia platyglossa            -0.0143 0.00761 566  -1.873
##  Bromus rubens - Phacelia tanacetifolia        0.0037 0.00624 566   0.594
##  Bromus rubens - Salvia columbariae           -0.0745 0.00785 566  -9.480
##  Layia platyglossa - Phacelia tanacetifolia    0.0180 0.00730 566   2.459
##  Layia platyglossa - Salvia columbariae       -0.0602 0.00854 566  -7.047
##  Phacelia tanacetifolia - Salvia columbariae  -0.0782 0.00756 566 -10.342
##  p.value
##   0.2409
##   0.9340
##   <.0001
##   0.0676
##   <.0001
##   <.0001
## 
## P value adjustment: tukey method for comparing a family of 4 estimates
model6 <- glm(succession ~ mean_temp*species + pendant_ID, family = "gaussian", data = final)
anova(model6, test = "Chisq")
## Analysis of Deviance Table
## 
## Model: gaussian, link: identity
## 
## Response: succession
## 
## Terms added sequentially (first to last)
## 
## 
##                   Df Deviance Resid. Df Resid. Dev  Pr(>Chi)    
## NULL                                702     360.37              
## mean_temp          1   20.933       701     339.44 1.300e-12 ***
## species            3   18.505       698     320.94 1.187e-09 ***
## pendant_ID         1    0.064       697     320.87    0.6956    
## mean_temp:species  3   32.219       694     288.65 < 2.2e-16 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
em6 <- emmeans(model6, pairwise~species|mean_temp)
em6
## $emmeans
## mean_temp = 26.3:
##  species                emmean     SE  df lower.CL upper.CL
##  Bromus rubens           0.677 0.0507 694    0.577    0.776
##  Layia platyglossa       0.691 0.0534 694    0.586    0.796
##  Phacelia tanacetifolia  0.999 0.0482 694    0.904    1.093
##  Salvia columbariae      0.642 0.0590 694    0.526    0.758
## 
## Confidence level used: 0.95 
## 
## $contrasts
## mean_temp = 26.3:
##  contrast                                    estimate     SE  df t.ratio
##  Bromus rubens - Layia platyglossa            -0.0144 0.0746 694  -0.193
##  Bromus rubens - Phacelia tanacetifolia       -0.3218 0.0694 694  -4.637
##  Bromus rubens - Salvia columbariae            0.0348 0.0780 694   0.447
##  Layia platyglossa - Phacelia tanacetifolia   -0.3074 0.0729 694  -4.219
##  Layia platyglossa - Salvia columbariae        0.0492 0.0792 694   0.621
##  Phacelia tanacetifolia - Salvia columbariae   0.3567 0.0763 694   4.672
##  p.value
##   0.9974
##   <.0001
##   0.9703
##   0.0002
##   0.9252
##   <.0001
## 
## P value adjustment: tukey method for comparing a family of 4 estimates
model7 <- glm(establishment~ mean_temp*species + pendant_ID, family = "gaussian", data = final)
anova(model7, test = "Chisq")
## Analysis of Deviance Table
## 
## Model: gaussian, link: identity
## 
## Response: establishment
## 
## Terms added sequentially (first to last)
## 
## 
##                   Df Deviance Resid. Df Resid. Dev Pr(>Chi)    
## NULL                                839     110890             
## mean_temp          1     1952       838     108938   <2e-16 ***
## species            3    94091       835      14847   <2e-16 ***
## pendant_ID         1        2       834      14845   0.7385    
## mean_temp:species  3     1361       831      13484   <2e-16 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
em7 <- emmeans(model7, pairwise~species|mean_temp)
em7
## $emmeans
## mean_temp = 26.5:
##  species                emmean    SE  df lower.CL upper.CL
##  Bromus rubens           4.966 0.294 831    4.389     5.54
##  Layia platyglossa       3.136 0.325 831    2.497     3.77
##  Phacelia tanacetifolia 27.615 0.296 831   27.035    28.20
##  Salvia columbariae      0.738 0.282 831    0.185     1.29
## 
## Confidence level used: 0.95 
## 
## $contrasts
## mean_temp = 26.5:
##  contrast                                    estimate    SE  df t.ratio p.value
##  Bromus rubens - Layia platyglossa               1.83 0.444 831   4.123  0.0002
##  Bromus rubens - Phacelia tanacetifolia        -22.65 0.413 831 -54.812  <.0001
##  Bromus rubens - Salvia columbariae              4.23 0.409 831  10.349  <.0001
##  Layia platyglossa - Phacelia tanacetifolia    -24.48 0.445 831 -54.999  <.0001
##  Layia platyglossa - Salvia columbariae          2.40 0.428 831   5.605  <.0001
##  Phacelia tanacetifolia - Salvia columbariae    26.88 0.410 831  65.582  <.0001
## 
## P value adjustment: tukey method for comparing a family of 4 estimates
model8 <- glm(germination ~ mean_temp*species + pendant_ID, family = "gaussian", data = final)
anova(model8, test = "Chisq")
## Analysis of Deviance Table
## 
## Model: gaussian, link: identity
## 
## Response: germination
## 
## Terms added sequentially (first to last)
## 
## 
##                   Df Deviance Resid. Df Resid. Dev  Pr(>Chi)    
## NULL                                839     105034              
## mean_temp          1       91       838     104944   0.05185 .  
## species            3    84141       835      20802 < 2.2e-16 ***
## pendant_ID         1       36       834      20766   0.22010    
## mean_temp:species  3      832       831      19934 1.408e-07 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
em8 <- emmeans(model8, pairwise~species|mean_temp)
em8
## $emmeans
## mean_temp = 26.5:
##  species                emmean    SE  df lower.CL upper.CL
##  Bromus rubens            6.54 0.357 831    5.839     7.24
##  Layia platyglossa        8.46 0.396 831    7.685     9.24
##  Phacelia tanacetifolia  27.69 0.359 831   26.988    28.40
##  Salvia columbariae       1.06 0.342 831    0.392     1.74
## 
## Confidence level used: 0.95 
## 
## $contrasts
## mean_temp = 26.5:
##  contrast                                    estimate    SE  df t.ratio p.value
##  Bromus rubens - Layia platyglossa              -1.92 0.540 831  -3.558  0.0022
##  Bromus rubens - Phacelia tanacetifolia        -21.15 0.502 831 -42.103  <.0001
##  Bromus rubens - Salvia columbariae              5.48 0.497 831  11.025  <.0001
##  Layia platyglossa - Phacelia tanacetifolia    -19.23 0.541 831 -35.540  <.0001
##  Layia platyglossa - Salvia columbariae          7.40 0.520 831  14.221  <.0001
##  Phacelia tanacetifolia - Salvia columbariae    26.63 0.498 831  53.443  <.0001
## 
## P value adjustment: tukey method for comparing a family of 4 estimates
###Test everything with Maximum Temp
#Max Temp by Average mass
ggplot(final, aes(max_temp, average_mass, color = temp),show.legend=FALSE) +
  geom_point() +
  facet_wrap(~species) +
  scale_color_brewer(palette = "Set1") + theme_classic() +
  geom_smooth(method = lm, se = TRUE) + 
  labs(x = "Maximum Temperature", y = "Bimass per Individual")
## `geom_smooth()` using formula = 'y ~ x'
## Warning: Removed 265 rows containing non-finite values (`stat_smooth()`).
## Warning: Removed 265 rows containing missing values (`geom_point()`).

model9 <- glm(average_mass ~ max_temp*species, family = "quasipoisson", data = final)
anova(model9, test = "Chisq")
## Analysis of Deviance Table
## 
## Model: quasipoisson, link: log
## 
## Response: average_mass
## 
## Terms added sequentially (first to last)
## 
## 
##                  Df Deviance Resid. Df Resid. Dev  Pr(>Chi)    
## NULL                               574     32.104              
## max_temp          1   0.3190       573     31.785 0.0153966 *  
## species           3   8.1628       570     23.622 < 2.2e-16 ***
## max_temp:species  3   0.9857       567     22.636 0.0004113 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
em9 <- emmeans(model9, pairwise~species|max_temp)
em9
## $emmeans
## max_temp = 28.2:
##  species                emmean    SE  df asymp.LCL asymp.UCL
##  Bromus rubens           -3.61 0.117 Inf     -3.84     -3.38
##  Layia platyglossa       -3.03 0.105 Inf     -3.23     -2.82
##  Phacelia tanacetifolia  -3.74 0.107 Inf     -3.95     -3.53
##  Salvia columbariae      -2.27 0.078 Inf     -2.42     -2.11
## 
## Results are given on the log (not the response) scale. 
## Confidence level used: 0.95 
## 
## $contrasts
## max_temp = 28.2:
##  contrast                                    estimate    SE  df z.ratio p.value
##  Bromus rubens - Layia platyglossa             -0.586 0.157 Inf  -3.733  0.0011
##  Bromus rubens - Phacelia tanacetifolia         0.125 0.159 Inf   0.788  0.8601
##  Bromus rubens - Salvia columbariae            -1.346 0.141 Inf  -9.574  <.0001
##  Layia platyglossa - Phacelia tanacetifolia     0.711 0.150 Inf   4.741  <.0001
##  Layia platyglossa - Salvia columbariae        -0.760 0.131 Inf  -5.818  <.0001
##  Phacelia tanacetifolia - Salvia columbariae   -1.471 0.133 Inf -11.084  <.0001
## 
## Results are given on the log (not the response) scale. 
## P value adjustment: tukey method for comparing a family of 4 estimates
###Succession of establishment (Established/germinated)
ggplot(final, aes(max_temp, succession, color = temp),show.legend=FALSE) +
  geom_point() +
  facet_wrap(~species) +
  scale_color_brewer(palette = "Set1") + theme_classic() +
  geom_smooth(method = lm, se = TRUE) + 
  labs(x = "Maximum Temperature", y = "Percent Successfully Established")
## `geom_smooth()` using formula = 'y ~ x'
## Warning: Removed 137 rows containing non-finite values (`stat_smooth()`).
## Warning: Removed 137 rows containing missing values (`geom_point()`).

model10 <- glm(succession ~ max_temp*species, family = "quasipoisson", data = final)
anova(model10, test = "Chisq")
## Analysis of Deviance Table
## 
## Model: quasipoisson, link: log
## 
## Response: succession
## 
## Terms added sequentially (first to last)
## 
## 
##                  Df Deviance Resid. Df Resid. Dev  Pr(>Chi)    
## NULL                               702     383.66              
## max_temp          1   22.861       701     360.80 4.069e-13 ***
## species           3   35.296       698     325.50 < 2.2e-16 ***
## max_temp:species  3   55.850       695     269.65 < 2.2e-16 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
em10 <- emmeans(model10, pairwise~species|max_temp)
em10
## $emmeans
## max_temp = 28.1:
##  species                  emmean     SE  df asymp.LCL asymp.UCL
##  Bromus rubens          -0.40300 0.0623 Inf   -0.5252   -0.2809
##  Layia platyglossa      -0.95327 0.0823 Inf   -1.1146   -0.7919
##  Phacelia tanacetifolia -0.00509 0.0466 Inf   -0.0964    0.0862
##  Salvia columbariae     -0.43995 0.0751 Inf   -0.5871   -0.2928
## 
## Results are given on the log (not the response) scale. 
## Confidence level used: 0.95 
## 
## $contrasts
## max_temp = 28.1:
##  contrast                                    estimate     SE  df z.ratio
##  Bromus rubens - Layia platyglossa             0.5503 0.1032 Inf   5.330
##  Bromus rubens - Phacelia tanacetifolia       -0.3979 0.0778 Inf  -5.115
##  Bromus rubens - Salvia columbariae            0.0369 0.0976 Inf   0.379
##  Layia platyglossa - Phacelia tanacetifolia   -0.9482 0.0946 Inf -10.026
##  Layia platyglossa - Salvia columbariae       -0.5133 0.1114 Inf  -4.607
##  Phacelia tanacetifolia - Salvia columbariae   0.4349 0.0884 Inf   4.921
##  p.value
##   <.0001
##   <.0001
##   0.9815
##   <.0001
##   <.0001
##   <.0001
## 
## Results are given on the log (not the response) scale. 
## P value adjustment: tukey method for comparing a family of 4 estimates
###Overall Establishment
ggplot(final, aes(max_temp, establishment, color = temp),show.legend=FALSE) +
  geom_point() +
  facet_wrap(~species) +
  scale_color_brewer(palette = "Set1") + theme_classic() +
  geom_smooth(method = lm, se = TRUE) + 
  labs(x = "Maximum Temperature", y = "Number Established Individuals")
## `geom_smooth()` using formula = 'y ~ x'

model11 <- glm(establishment ~ max_temp*species, family = "quasipoisson", data = final)
anova(model11, test = "Chisq")
## Analysis of Deviance Table
## 
## Model: quasipoisson, link: log
## 
## Response: establishment
## 
## Terms added sequentially (first to last)
## 
## 
##                  Df Deviance Resid. Df Resid. Dev Pr(>Chi)    
## NULL                               839    12268.9             
## max_temp          1      2.1       838    12266.8   0.4024    
## species           3   9621.6       835     2645.3   <2e-16 ***
## max_temp:species  3    319.3       832     2326.0   <2e-16 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
em11 <- emmeans(model11, pairwise~species|max_temp)
em11
## $emmeans
## max_temp = 28.4:
##  species                emmean     SE  df asymp.LCL asymp.UCL
##  Bromus rubens           1.464 0.0605 Inf     1.346    1.5827
##  Layia platyglossa       0.978 0.0749 Inf     0.831    1.1251
##  Phacelia tanacetifolia  3.302 0.0231 Inf     3.257    3.3475
##  Salvia columbariae     -0.324 0.1452 Inf    -0.609   -0.0396
## 
## Results are given on the log (not the response) scale. 
## Confidence level used: 0.95 
## 
## $contrasts
## max_temp = 28.4:
##  contrast                                    estimate     SE  df z.ratio
##  Bromus rubens - Layia platyglossa              0.486 0.0963 Inf   5.045
##  Bromus rubens - Phacelia tanacetifolia        -1.838 0.0648 Inf -28.376
##  Bromus rubens - Salvia columbariae             1.788 0.1573 Inf  11.369
##  Layia platyglossa - Phacelia tanacetifolia    -2.324 0.0784 Inf -29.632
##  Layia platyglossa - Salvia columbariae         1.302 0.1634 Inf   7.971
##  Phacelia tanacetifolia - Salvia columbariae    3.626 0.1470 Inf  24.663
##  p.value
##   <.0001
##   <.0001
##   <.0001
##   <.0001
##   <.0001
##   <.0001
## 
## Results are given on the log (not the response) scale. 
## P value adjustment: tukey method for comparing a family of 4 estimates
###Overall Germination

ggplot(final, aes(max_temp, germination, color = temp),show.legend=FALSE) +
  geom_point() +
  facet_wrap(~species) +
  scale_color_brewer(palette = "Set1") + theme_classic() +
  geom_smooth(method = lm, se = TRUE) + 
  labs(x = "Maximum Temperature", y = "Number Germinated Individuals")
## `geom_smooth()` using formula = 'y ~ x'

model12 <- glm(germination ~ temp*species, family = "quasipoisson", data = final)
anova(model12, test = "Chisq")
## Analysis of Deviance Table
## 
## Model: quasipoisson, link: log
## 
## Response: germination
## 
## Terms added sequentially (first to last)
## 
## 
##              Df Deviance Resid. Df Resid. Dev  Pr(>Chi)    
## NULL                           839    10186.2              
## temp          2    275.3       837     9910.9 < 2.2e-16 ***
## species       3   7283.5       834     2627.4 < 2.2e-16 ***
## temp:species  6    195.9       828     2431.5 1.548e-12 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
em12 <- emmeans(model12, pairwise~temp|species)
em12
## $emmeans
## species = Bromus rubens:
##  temp   emmean     SE  df asymp.LCL asymp.UCL
##  High    1.035 0.1217 Inf     0.796     1.273
##  Low     2.253 0.0662 Inf     2.123     2.382
##  Medium  1.609 0.0913 Inf     1.431     1.788
## 
## species = Layia platyglossa:
##  temp   emmean     SE  df asymp.LCL asymp.UCL
##  High    1.870 0.0802 Inf     1.713     2.027
##  Low     2.496 0.0586 Inf     2.381     2.610
##  Medium  2.422 0.0608 Inf     2.303     2.541
## 
## species = Phacelia tanacetifolia:
##  temp   emmean     SE  df asymp.LCL asymp.UCL
##  High    3.162 0.0420 Inf     3.080     3.244
##  Low     3.361 0.0380 Inf     3.286     3.435
##  Medium  3.361 0.0380 Inf     3.287     3.436
## 
## species = Salvia columbariae:
##  temp   emmean     SE  df asymp.LCL asymp.UCL
##  High   -0.138 0.2187 Inf    -0.566     0.291
##  Low     0.121 0.1921 Inf    -0.256     0.498
##  Medium  0.158 0.1886 Inf    -0.211     0.528
## 
## Results are given on the log (not the response) scale. 
## Confidence level used: 0.95 
## 
## $contrasts
## species = Bromus rubens:
##  contrast       estimate     SE  df z.ratio p.value
##  High - Low    -1.218086 0.1385 Inf  -8.795  <.0001
##  High - Medium -0.574729 0.1521 Inf  -3.779  0.0005
##  Low - Medium   0.643356 0.1128 Inf   5.706  <.0001
## 
## species = Layia platyglossa:
##  contrast       estimate     SE  df z.ratio p.value
##  High - Low    -0.625962 0.0993 Inf  -6.304  <.0001
##  High - Medium -0.552669 0.1006 Inf  -5.494  <.0001
##  Low - Medium   0.073293 0.0845 Inf   0.868  0.6606
## 
## species = Phacelia tanacetifolia:
##  contrast       estimate     SE  df z.ratio p.value
##  High - Low    -0.199019 0.0567 Inf  -3.512  0.0013
##  High - Medium -0.199515 0.0567 Inf  -3.522  0.0012
##  Low - Medium  -0.000496 0.0538 Inf  -0.009  1.0000
## 
## species = Salvia columbariae:
##  contrast       estimate     SE  df z.ratio p.value
##  High - Low    -0.258574 0.2911 Inf  -0.888  0.6477
##  High - Medium -0.295845 0.2888 Inf  -1.025  0.5613
##  Low - Medium  -0.037271 0.2692 Inf  -0.138  0.9895
## 
## Results are given on the log (not the response) scale. 
## P value adjustment: tukey method for comparing a family of 3 estimates
###Test establish density by Biomass, Temp Block, and Species

#Based on number of individuals that survived germination
ggplot(final, aes(mean_temp, succession, color = temp),show.legend=FALSE) +
  geom_point() +
  facet_wrap(~species) +
  scale_color_brewer(palette = "Set1") + theme_classic() +
  geom_smooth(method = lm, se = TRUE) + 
  labs(x = "Temperature", y = "Number Germinated Individuals")
## `geom_smooth()` using formula = 'y ~ x'
## Warning: Removed 137 rows containing non-finite values (`stat_smooth()`).
## Warning: Removed 137 rows containing missing values (`geom_point()`).

model13 <- glm(succession ~ mass*temp*species, family = "quasipoisson", data = final)
anova(model13, test = "Chisq")
## Analysis of Deviance Table
## 
## Model: quasipoisson, link: log
## 
## Response: succession
## 
## Terms added sequentially (first to last)
## 
## 
##                   Df Deviance Resid. Df Resid. Dev  Pr(>Chi)    
## NULL                                702     383.66              
## mass               1   27.714       701     355.95 7.811e-14 ***
## temp               2    5.471       699     350.47 0.0040343 ** 
## species            3   20.344       696     330.13 6.543e-09 ***
## mass:temp          2    8.868       694     321.26 0.0001316 ***
## mass:species       3   11.946       691     309.32 2.410e-05 ***
## temp:species       6   51.170       685     258.15 < 2.2e-16 ***
## mass:temp:species  6    5.959       679     252.19 0.0617671 .  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
em13 <- emmeans(model13, pairwise~temp|species)
## NOTE: Results may be misleading due to involvement in interactions
em13
## $emmeans
## species = Bromus rubens:
##  temp     emmean    SE  df asymp.LCL asymp.UCL
##  High    0.70130 0.589 Inf    -0.454    1.8566
##  Low    -0.20670 0.140 Inf    -0.482    0.0684
##  Medium -0.20447 0.180 Inf    -0.558    0.1488
## 
## species = Layia platyglossa:
##  temp     emmean    SE  df asymp.LCL asymp.UCL
##  High   -0.11708 0.137 Inf    -0.386    0.1523
##  Low    -1.90094 0.583 Inf    -3.044   -0.7574
##  Medium -1.05329 0.191 Inf    -1.427   -0.6792
## 
## species = Phacelia tanacetifolia:
##  temp     emmean    SE  df asymp.LCL asymp.UCL
##  High   -0.00536 0.124 Inf    -0.248    0.2369
##  Low    -0.00089 0.155 Inf    -0.305    0.3028
##  Medium -0.00290 0.163 Inf    -0.323    0.3172
## 
## species = Salvia columbariae:
##  temp     emmean    SE  df asymp.LCL asymp.UCL
##  High    0.23943 0.204 Inf    -0.160    0.6393
##  Low     0.03293 0.235 Inf    -0.428    0.4938
##  Medium -0.02247 0.179 Inf    -0.374    0.3287
## 
## Results are given on the log (not the response) scale. 
## Confidence level used: 0.95 
## 
## $contrasts
## species = Bromus rubens:
##  contrast      estimate    SE  df z.ratio p.value
##  High - Low     0.90800 0.606 Inf   1.499  0.2916
##  High - Medium  0.90577 0.616 Inf   1.470  0.3057
##  Low - Medium  -0.00223 0.228 Inf  -0.010  0.9999
## 
## species = Layia platyglossa:
##  contrast      estimate    SE  df z.ratio p.value
##  High - Low     1.78387 0.599 Inf   2.976  0.0082
##  High - Medium  0.93621 0.235 Inf   3.981  0.0002
##  Low - Medium  -0.84765 0.614 Inf  -1.381  0.3510
## 
## species = Phacelia tanacetifolia:
##  contrast      estimate    SE  df z.ratio p.value
##  High - Low    -0.00447 0.198 Inf  -0.023  0.9997
##  High - Medium -0.00246 0.205 Inf  -0.012  0.9999
##  Low - Medium   0.00201 0.225 Inf   0.009  1.0000
## 
## species = Salvia columbariae:
##  contrast      estimate    SE  df z.ratio p.value
##  High - Low     0.20649 0.311 Inf   0.663  0.7848
##  High - Medium  0.26190 0.272 Inf   0.965  0.5993
##  Low - Medium   0.05541 0.296 Inf   0.187  0.9808
## 
## Results are given on the log (not the response) scale. 
## P value adjustment: tukey method for comparing a family of 3 estimates

###Ideas from Previous version Purpose: The purpose of this experiment is to determine if cooler temperatures, as seen under shrub canopies, could be used as an indicator for the germination success of both native and exotic plant species in arid ecosystems.

Hypothesis: We hypothesize that variations in fine scale temperature can act as a direct proxy for successful desert plant establishment and germination.

Predictions: 1) Fine scale temperature can be experimentally manipulated via heat lamps in an enclosed setting. 2) Plants will respond to varying fine scale temperatures. 3) The response to temperature will be species specific. 4) Responses to temperature will vary between native and exotic plant species.

Data: All data can be access on KNB. https://knb.ecoinformatics.org/view/doi:10.5063/F1GQ6W6R

Rough Ideas: 1)To connect this to shrub density I want to show that the cooler temperatures, as experienced under shrubs, can provide a benefit for species germination. The law table is meant to simulate these lower temperatures while the medium and high tables are meant to act as open areas at a moderate temperature and at extreme temperature. 2)There should be a visible difference between overall germination and mass between native and invasive species. If we can determine if invasives can germinate better under shrubs than in open areas then we can connect this to competition between natives and invasive. 3)Species should vary in their response to fine scale temperature since the range that they germinate vary from species to species. We should see that some natives germinate better at specific temperature while others do not. 4) We can connect this possibly to climate change. If we find that higher temperatures means a lower germination success then that can be the take-home message, that increasing temperatures have negative effects on desert plant species.