Analysis of field and evironmental data- April
Chelsey Wegener
4/8/2021
What’s new since last meeting (focused on research questions 1 and 2 (listed below) so these modifications only apply to those):
-month.yr is now a date factor. Since month-year is not a format recognized by R I added “01” as the date for all of them. This should be ok since we’re more interested in month and year than date anyway. I like that it’s a date vs a character now
-Research question 1: Site is fixed factor in all lm
-Research question 1:tried month-year as fixed and random. When it’s random I added an interaction term between them
-Research question 2: tried season as fixed factor and site as a random factor (to account for the fact they spatially aggregated) to see if they are all responding in the same way over time
-Research question 2: keeping season as time factor for now but can also try it as month-yr
-Research question 2: looked at model without interaction term as well
Things to consider:
-fixed vs random sites. Doesn’t have to be apriori. Are the sites different? or are they spatial blocks?
Note for self (sing I keep forgetting): fixed factor use factor(x/factor of interest), for random factor just list column. For example, in lm(cover ~ factor(site) + month.yr, data =all), site is fized and month.yr is random
Questions for analysis:
ABUNDANCE (site only, no time)
1. Is there a difference in Fucus abundance between sites?
1.1 Is there a difference in Fucus percent cover amoung sites?
1.2 Is there a difference in Fucus total density amoung sites?
notes: include sampling date as random factor? Samples independent of each other? Time fixed or random? Find best fit (Gaussain, Poisson (density), normal). Percent cover as a proporation (i.e. 50% = 0.5)
-Site is the main factor. Sample month-year is a random factor but try both. Site is definitely a fixed factor.
-when site and time are fixed, add an interaction term to assess model fit.
ABUNDANCE (site and time)
2. Is there a difference in Fucus abundance between site and time? (site:time)
2.1 Is there a difference in Fucus percent cover between site and time?
2.2 Is there a difference in Fucus total density between site and time?
notes: Find best fit (Gaussain, Poisson (density), normal). Seasonality time by quadrat. Percent cover as a proporation (i.e. 50% = 0.5)
-Seasonaly: time as fixed factor, site as a random factor (to account for the fact they spatially aggregated). –> are they all responding in the same way over time?
REPRODUCTIVE PHENOLOGY (with seasonality)
3. Is reproduction affected by the season? 3.1 Is oogonia/receptacle affected by the season?
3.2 Is oogonia/thalli affected by the season?
3.3 Is conceptical/thalli affected by the season?
3.4 Is percent reproductive tissue affected by the season?
3.5 Is adult density affected by the season?
3.6 Is juvenile density affected by the season?
REPRODUCTIVE PHENOLOGY (without seasonality)
4. Is reproductive phenology affected by environmental conditions (salinity, pH, water temperature and air temperature)? (not sure which metric of each variable I should be using? mean? daily min?) For now I’m just using mean between survey dates which is a good starting point while I’m figuring out the code.
4.1 Is oogoina/receptical affected by environmental conditions (salinity, pH, water temperature and air temperature)?
4.2 Is oogonia/thalli affected by environmental conditions (salinity, pH, water temperature and air temperature)?
4.3 Is conceptacle/thalli affected by environmental conditions (salinity, pH, water temperature and air temperature)?
4.4 Is percent reproductive tissue affected by environmental conditions (salinity, pH, water temperature and air temperature)?
4.5 Is juvenile density affected by environmental conditions (salinity, pH, water temperature and air temperature)?
4.6 Is adult density affected by environmental conditions (salinity, pH, water temperature and air temperature)?
notes: site and time as random factors. x ~ salintiy + ph + water temp + air temp + site(random) + time(random). In AIC ranking, do a model for every combination.
Additional inquires:
-timing of oogonia/thalli -> juvenile abundance
Notes:
-Time: month + year or season +year or season:year –> I don’t think I’m doing this correctly right now. I have a season year column and a month year column. I think it accomplishes the same thing but I should probably have these columns all separate and then add an iteraction term. So far season shows more of an effect than month so I’ll be using season-year for all “time” factors" in this markdown (at least for now).
Initial interpretation of results:
1. Is there a difference in Fucus abundance between sites?
Looks like there is possibly a difference in Fucus cover and density between sites with P ~= 0.06 - 0.07
2. Is there a difference in Fucus abundance between site and time? (site:time)
There is a difference in Fucus cover P <.001 and density between site and time
3. Is reproduction affected by the season?
Only number of large thalli and number of small thalli are affected by season. Other reproductive measures didn’t show significance
4. Is reproductive phenology affected by environmental conditions (salinity, pH, water temperature and air temperature)? Water temperature affects oogonia per receptical
Water temperature and salinity affects the number of small thalli per quadrat
Set up
rm(list=ls())
library(tidyverse)
library(ggpubr)
library(scales)
library(chron)
library(plotly)
library(taRifx)
library(aweek)
library(easypackages)
library(renv)
library(here)
library(ggthemes)
library(gridExtra)
library(patchwork)
library(tidyquant)
library(recipes)
library(cranlogs)
library(knitr)
library(openair)
#new
library(AICcmodavg)
library(plm) #for random factors
library(lme4)Read in data
#read in data
all<-read.csv(
"https://raw.githubusercontent.com/Cmwegener/thesis/master/data/environment_field/envi.field.all.csv",
header = TRUE
)Make month year column. Was originally going to do just month but since I have some duplicate months I decided to do month-year. NOTE: since you need a day to convert to date format, I added the same date to all of them so that the only difference is the month-year which is what we’re really interested in anyway
all$date<-as.Date(all$date, format = c("%Y-%m-%d"))
all$month.yr<-format(all$date, "%Y-%m") #can't convert month-year format as date so it's character, I'm going to assign them all the same date then so that it can be converted to a date
all$day<-"-01" #assigning place holder "day"
all$month.yr<-paste(all$month.yr,all$day, sep="")#merging with month.yr column
all$month.yr<-as.Date(all$month.yr, format=c("%Y-%m-%d")) #format as date
all<-subset(all, select = -c(day) ) #remove unneeded day column
#probably a much cleaner way to do this but this worksSeason column. Split year into quarters. Added year since there’s some duplicate months? Since it’s a character I’ll need to list all the month.yr per season.
2018-06 - 2018-08 summer.2018
2018-09 - 2018-11 fall.2018
2018-12 - 2019-02 winter.2018
2019-03 - 2019-05 spring.2019
2019-06 - 2019-08 summer.2019
2019-09 - 2019-11 fall.2019 incomplete season, only have field data for 2019-09 –> Think I should probably remove this term and put “NA” since I only have field data for that month. Doesn’t seem right to compare it to the other seasons with more months and data.
all$season[all$month.yr == "2018-06-01"] <- "summer.2018"
all$season[all$month.yr == "2018-07-01"] <- "summer.2018"
all$season[all$month.yr == "2018-08-01"] <- "summer.2018"
all$season[all$month.yr == "2018-09-01"] <- "fall.2018"
all$season[all$month.yr == "2018-10-01"] <- "fall.2018"
all$season[all$month.yr == "2018-11-01"] <- "fall.2018"
all$season[all$month.yr == "2018-12-01"] <- "winter.2018"
all$season[all$month.yr == "2019-01-01"] <- "winter.2018"
all$season[all$month.yr == "2019-02-01"] <- "winter.2018"
all$season[all$month.yr == "2019-03-01"] <- "spring.2019"
all$season[all$month.yr == "2019-04-01"] <- "spring.2019"
all$season[all$month.yr == "2019-05-01"] <- "spring.2019"
all$season[all$month.yr == "2019-06-01"] <- "summer.2019"
all$season[all$month.yr == "2019-07-01"] <- "summer.2019"
all$season[all$month.yr == "2019-08-01"] <- "summer.2019"
all$season[all$month.yr == "2019-09-01"] <- "fall.2019"Converting percent cover as a proporation (i.e. 50% = 0.5) to make it easier to interpret
all$cover<-all$cover/100####ABUNDANCE (site only, no time)####
1. Is there a difference in Fucus abundance between sites?
1.1 Is there a difference in Fucus percent cover amoung sites?
1.2 Is there a difference in Fucus total density amoung sites?
notes: include sampling date as random factor? Samples independent of each other? Time fixed or random? Percent cover as a proporation (i.e. 50% = 0.5).
Looking at site as fixed and random
Notes from most recent meeting:
-Site is the main factor. Sample month-year is a random factor but try both. Site is definitely a fixed factor.
-when site and time are fixed, add an interaction term to assess model fit.
####1.1 Is there a difference in Fucus percent cover amoung sites?####
percent cover ~ Site as fixed factor + month-year as random
lm1.1 <- lm(cover ~ factor(site) + month.yr, data =all)
lm1.1##
## Call:
## lm(formula = cover ~ factor(site) + month.yr, data = all)
##
## Coefficients:
## (Intercept) factor(site)hs.fp factor(site)nd.eos factor(site)pc.cc
## 8.8468653 -0.1084375 -0.0878202 -0.0114250
## month.yr
## -0.0004728summary (lm1.1)##
## Call:
## lm(formula = cover ~ factor(site) + month.yr, data = all)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.23229 -0.07419 0.01286 0.08046 0.21931
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 8.8468653 1.8386227 4.812 1.17e-05 ***
## factor(site)hs.fp -0.1084375 0.0402736 -2.693 0.00934 **
## factor(site)nd.eos -0.0878202 0.0409426 -2.145 0.03631 *
## factor(site)pc.cc -0.0114250 0.0417085 -0.274 0.78515
## month.yr -0.0004728 0.0001026 -4.606 2.41e-05 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.1139 on 56 degrees of freedom
## Multiple R-squared: 0.3593, Adjusted R-squared: 0.3136
## F-statistic: 7.853 on 4 and 56 DF, p-value: 4.259e-05anova (lm1.1)## Analysis of Variance Table
##
## Response: cover
## Df Sum Sq Mean Sq F value Pr(>F)
## factor(site) 3 0.13228 0.044095 3.3983 0.02389 *
## month.yr 1 0.27529 0.275291 21.2159 2.41e-05 ***
## Residuals 56 0.72664 0.012976
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1plot (lm1.1)
both factors look significant, and model fit looks good enough
percent cover ~ Site as fixed factor + month-year as fixed factor + interaction term
lm1.1 <- lm(cover ~ factor(site) + factor(month.yr) + site:month.yr, data =all)
lm1.1##
## Call:
## lm(formula = cover ~ factor(site) + factor(month.yr) + site:month.yr,
## data = all)
##
## Coefficients:
## (Intercept) factor(site)hs.fp
## 5.040e-01 -1.471e+01
## factor(site)nd.eos factor(site)pc.cc
## -3.604e+00 -7.976e-02
## factor(month.yr)2018-07-01 factor(month.yr)2018-08-01
## 2.176e-02 5.427e-02
## factor(month.yr)2018-09-01 factor(month.yr)2018-10-01
## 8.879e-02 7.847e-02
## factor(month.yr)2018-11-01 factor(month.yr)2018-12-01
## 8.808e-02 1.382e-02
## factor(month.yr)2019-01-01 factor(month.yr)2019-02-01
## 1.658e-02 -2.565e-02
## factor(month.yr)2019-03-01 factor(month.yr)2019-04-01
## -1.439e-02 -3.013e-02
## factor(month.yr)2019-05-01 factor(month.yr)2019-06-01
## 6.132e-03 -1.706e-01
## factor(month.yr)2019-07-01 factor(month.yr)2019-08-01
## -2.773e-01 -2.866e-01
## factor(month.yr)2019-09-01 siteby.rb:month.yr
## -2.746e-01 -4.491e-06
## sitehs.fp:month.yr sitend.eos:month.yr
## 8.108e-04 1.921e-04
## sitepc.cc:month.yr
## NAsummary (lm1.1)##
## Call:
## lm(formula = cover ~ factor(site) + factor(month.yr) + site:month.yr,
## data = all)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.146970 -0.048865 -0.005254 0.054135 0.116080
##
## Coefficients: (1 not defined because of singularities)
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 5.040e-01 3.774e+00 0.134 0.894450
## factor(site)hs.fp -1.471e+01 3.714e+00 -3.961 0.000308 ***
## factor(site)nd.eos -3.604e+00 3.742e+00 -0.963 0.341508
## factor(site)pc.cc -7.976e-02 3.800e+00 -0.021 0.983360
## factor(month.yr)2018-07-01 2.176e-02 5.826e-02 0.374 0.710771
## factor(month.yr)2018-08-01 5.427e-02 5.867e-02 0.925 0.360644
## factor(month.yr)2018-09-01 8.879e-02 5.936e-02 1.496 0.142799
## factor(month.yr)2018-10-01 7.847e-02 6.402e-02 1.226 0.227650
## factor(month.yr)2018-11-01 8.808e-02 7.395e-02 1.191 0.240827
## factor(month.yr)2018-12-01 1.382e-02 6.288e-02 0.220 0.827159
## factor(month.yr)2019-01-01 1.658e-02 6.454e-02 0.257 0.798557
## factor(month.yr)2019-02-01 -2.565e-02 6.641e-02 -0.386 0.701383
## factor(month.yr)2019-03-01 -1.439e-02 6.826e-02 -0.211 0.834111
## factor(month.yr)2019-04-01 -3.013e-02 7.047e-02 -0.428 0.671352
## factor(month.yr)2019-05-01 6.132e-03 7.277e-02 0.084 0.933272
## factor(month.yr)2019-06-01 -1.706e-01 7.529e-02 -2.266 0.029072 *
## factor(month.yr)2019-07-01 -2.773e-01 7.784e-02 -3.563 0.000987 ***
## factor(month.yr)2019-08-01 -2.866e-01 8.060e-02 -3.555 0.001008 **
## factor(month.yr)2019-09-01 -2.746e-01 8.347e-02 -3.289 0.002134 **
## siteby.rb:month.yr -4.491e-06 2.120e-04 -0.021 0.983208
## sitehs.fp:month.yr 8.108e-04 2.120e-04 3.824 0.000462 ***
## sitend.eos:month.yr 1.921e-04 2.125e-04 0.904 0.371560
## sitepc.cc:month.yr NA NA NA NA
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.0822 on 39 degrees of freedom
## Multiple R-squared: 0.7677, Adjusted R-squared: 0.6425
## F-statistic: 6.136 on 21 and 39 DF, p-value: 5.918e-07anova (lm1.1)## Analysis of Variance Table
##
## Response: cover
## Df Sum Sq Mean Sq F value Pr(>F)
## factor(site) 3 0.13228 0.044095 6.5255 0.0011031 **
## factor(month.yr) 15 0.60030 0.040020 5.9225 3.821e-06 ***
## site:month.yr 3 0.13810 0.046033 6.8124 0.0008404 ***
## Residuals 39 0.26353 0.006757
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1plot (lm1.1)
Both significant, not sure about Residuals vs Leverage plot
####1.2 Is there a difference in Fucus total density amoung sites?###
density ~ site as fixed factor + month-year as random
lm1.2 <- lm(no.fuc.q ~ factor(site) + month.yr, data =all)
lm1.2##
## Call:
## lm(formula = no.fuc.q ~ factor(site) + month.yr, data = all)
##
## Coefficients:
## (Intercept) factor(site)hs.fp factor(site)nd.eos factor(site)pc.cc
## 2245.5084 -30.3625 -9.0141 -2.9728
## month.yr
## -0.1227summary (lm1.2)##
## Call:
## lm(formula = no.fuc.q ~ factor(site) + month.yr, data = all)
##
## Residuals:
## Min 1Q Median 3Q Max
## -43.583 -19.693 -4.829 13.943 97.399
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2245.5084 454.9479 4.936 7.54e-06 ***
## factor(site)hs.fp -30.3625 9.9653 -3.047 0.00352 **
## factor(site)nd.eos -9.0141 10.1308 -0.890 0.37740
## factor(site)pc.cc -2.9728 10.3203 -0.288 0.77437
## month.yr -0.1227 0.0254 -4.830 1.10e-05 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 28.19 on 56 degrees of freedom
## Multiple R-squared: 0.3782, Adjusted R-squared: 0.3338
## F-statistic: 8.517 on 4 and 56 DF, p-value: 1.93e-05anova (lm1.2)## Analysis of Variance Table
##
## Response: no.fuc.q
## Df Sum Sq Mean Sq F value Pr(>F)
## factor(site) 3 8534 2844.5 3.5805 0.01933 *
## month.yr 1 18531 18530.9 23.3253 1.099e-05 ***
## Residuals 56 44490 794.5
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1plot (lm1.2)
Both factors significant, plots looks good but not sure about Residuals vs Leverage plot
density ~ site as fixed factor + month.yr as fixed factor + interaction term
lm1.2 <- lm(no.fuc.q ~ factor(site) + factor(month.yr) + site:month.yr, data =all)
lm1.2##
## Call:
## lm(formula = no.fuc.q ~ factor(site) + factor(month.yr) + site:month.yr,
## data = all)
##
## Coefficients:
## (Intercept) factor(site)hs.fp
## -1.420e+03 -1.807e+03
## factor(site)nd.eos factor(site)pc.cc
## 4.902e+02 1.481e+03
## factor(month.yr)2018-07-01 factor(month.yr)2018-08-01
## 4.568e+01 6.612e+01
## factor(month.yr)2018-09-01 factor(month.yr)2018-10-01
## 3.494e+01 -7.665e-01
## factor(month.yr)2018-11-01 factor(month.yr)2018-12-01
## -1.631e+01 -2.769e+01
## factor(month.yr)2019-01-01 factor(month.yr)2019-02-01
## -2.959e+01 -3.357e+01
## factor(month.yr)2019-03-01 factor(month.yr)2019-04-01
## -3.756e+01 -3.801e+01
## factor(month.yr)2019-05-01 factor(month.yr)2019-06-01
## -3.309e+01 -5.189e+01
## factor(month.yr)2019-07-01 factor(month.yr)2019-08-01
## -3.937e+01 -4.147e+01
## factor(month.yr)2019-09-01 siteby.rb:month.yr
## -4.763e+01 8.288e-02
## sitehs.fp:month.yr sitend.eos:month.yr
## 1.821e-01 5.495e-02
## sitepc.cc:month.yr
## NAsummary (lm1.2)##
## Call:
## lm(formula = no.fuc.q ~ factor(site) + factor(month.yr) + site:month.yr,
## data = all)
##
## Residuals:
## Min 1Q Median 3Q Max
## -43.963 -10.954 -1.371 9.329 36.748
##
## Coefficients: (1 not defined because of singularities)
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -1.420e+03 8.785e+02 -1.617 0.113995
## factor(site)hs.fp -1.807e+03 8.646e+02 -2.090 0.043157 *
## factor(site)nd.eos 4.902e+02 8.712e+02 0.563 0.576835
## factor(site)pc.cc 1.481e+03 8.845e+02 1.674 0.102080
## factor(month.yr)2018-07-01 4.568e+01 1.356e+01 3.368 0.001714 **
## factor(month.yr)2018-08-01 6.612e+01 1.366e+01 4.841 2.07e-05 ***
## factor(month.yr)2018-09-01 3.494e+01 1.382e+01 2.529 0.015612 *
## factor(month.yr)2018-10-01 -7.665e-01 1.490e+01 -0.051 0.959239
## factor(month.yr)2018-11-01 -1.631e+01 1.722e+01 -0.948 0.349130
## factor(month.yr)2018-12-01 -2.769e+01 1.464e+01 -1.891 0.066013 .
## factor(month.yr)2019-01-01 -2.959e+01 1.502e+01 -1.970 0.056025 .
## factor(month.yr)2019-02-01 -3.357e+01 1.546e+01 -2.172 0.036033 *
## factor(month.yr)2019-03-01 -3.756e+01 1.589e+01 -2.364 0.023168 *
## factor(month.yr)2019-04-01 -3.801e+01 1.641e+01 -2.317 0.025836 *
## factor(month.yr)2019-05-01 -3.309e+01 1.694e+01 -1.953 0.057989 .
## factor(month.yr)2019-06-01 -5.189e+01 1.753e+01 -2.961 0.005197 **
## factor(month.yr)2019-07-01 -3.937e+01 1.812e+01 -2.172 0.035962 *
## factor(month.yr)2019-08-01 -4.147e+01 1.876e+01 -2.210 0.033026 *
## factor(month.yr)2019-09-01 -4.763e+01 1.943e+01 -2.451 0.018830 *
## siteby.rb:month.yr 8.288e-02 4.936e-02 1.679 0.101108
## sitehs.fp:month.yr 1.821e-01 4.936e-02 3.689 0.000685 ***
## sitend.eos:month.yr 5.495e-02 4.947e-02 1.111 0.273467
## sitepc.cc:month.yr NA NA NA NA
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 19.14 on 39 degrees of freedom
## Multiple R-squared: 0.8004, Adjusted R-squared: 0.693
## F-statistic: 7.448 on 21 and 39 DF, p-value: 4.419e-08anova (lm1.2)## Analysis of Variance Table
##
## Response: no.fuc.q
## Df Sum Sq Mean Sq F value Pr(>F)
## factor(site) 3 8534 2844.51 7.7683 0.0003479 ***
## factor(month.yr) 15 43443 2896.17 7.9094 1.038e-07 ***
## site:month.yr 3 5297 1765.78 4.8223 0.0059722 **
## Residuals 39 14281 366.17
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1plot (lm1.2)
Got warning: “ANOVA F-tests on an essentially perfect fit are unreliable” need to look into more and couldn’t get plot. Might have to do with site and month.yr being character?
####ABUNDANCE (site and time)####
2. Is there a difference in Fucus abundance between site and time? (site:time)
2.1 Is there a difference in Fucus percent cover between site and time?
2.2 Is there a difference in Fucus total density between site and time?
notes:Find best fit (Gaussain, Poisson (density), normal). Seasonality time by quadrat. Percent cover as a proporation (i.e. 50% = 0.5)
From last meeting: Seasonaly: time as fixed factor, site as a random factor (to account for the fact they spatially aggregated). –> are they all responding in the same way over time?
####2.1 Is there a difference in Fucus percent cover between site and time?####
cover ~ site as random factor + season as fixed factor
lm2.1 <- lm(cover ~ site + factor(season), data =all)
lm2.1##
## Call:
## lm(formula = cover ~ site + factor(season), data = all)
##
## Coefficients:
## (Intercept) sitehs.fp
## 0.475838 -0.108437
## sitend.eos sitepc.cc
## -0.083744 -0.003737
## factor(season)fall.2019 factor(season)spring.2019
## -0.268358 -0.044858
## factor(season)summer.2018 factor(season)summer.2019
## -0.074941 -0.254024
## factor(season)winter.2018
## -0.052858summary (lm2.1)##
## Call:
## lm(formula = cover ~ site + factor(season), data = all)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.15554 -0.06346 0.00102 0.06246 0.13916
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.475838 0.035849 13.274 < 2e-16 ***
## sitehs.fp -0.108437 0.032715 -3.315 0.00168 **
## sitend.eos -0.083744 0.033306 -2.514 0.01506 *
## sitepc.cc -0.003737 0.034089 -0.110 0.91312
## factor(season)fall.2019 -0.268358 0.055785 -4.811 1.33e-05 ***
## factor(season)spring.2019 -0.044858 0.041048 -1.093 0.27951
## factor(season)summer.2018 -0.074941 0.041048 -1.826 0.07364 .
## factor(season)summer.2019 -0.254024 0.041048 -6.188 9.69e-08 ***
## factor(season)winter.2018 -0.052858 0.041048 -1.288 0.20355
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.09253 on 52 degrees of freedom
## Multiple R-squared: 0.6074, Adjusted R-squared: 0.5471
## F-statistic: 10.06 on 8 and 52 DF, p-value: 2.433e-08anova (lm2.1)## Analysis of Variance Table
##
## Response: cover
## Df Sum Sq Mean Sq F value Pr(>F)
## site 3 0.13228 0.044095 5.1499 0.003417 **
## factor(season) 5 0.55669 0.111338 13.0032 3.216e-08 ***
## Residuals 52 0.44524 0.008562
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1plot (lm2.1)
Both significant, plots look good except not sure about the residual vs leverage one
cover ~ site as random factor + season as fixed factor + interaction term
lm2.1 <- lm(cover ~ site + season + site:season, data =all)
lm2.1##
## Call:
## lm(formula = cover ~ site + season + site:season, data = all)
##
## Coefficients:
## (Intercept) sitehs.fp
## 0.538000 -0.251667
## sitend.eos sitepc.cc
## -0.172500 0.044000
## seasonfall.2019 seasonspring.2019
## -0.397000 -0.083333
## seasonsummer.2018 seasonsummer.2019
## -0.135000 -0.408667
## seasonwinter.2018 sitehs.fp:seasonfall.2019
## -0.088333 0.268667
## sitend.eos:seasonfall.2019 sitepc.cc:seasonfall.2019
## 0.201500 -0.020000
## sitehs.fp:seasonspring.2019 sitend.eos:seasonspring.2019
## 0.197000 0.004833
## sitepc.cc:seasonspring.2019 sitehs.fp:seasonsummer.2018
## -0.112333 0.059333
## sitend.eos:seasonsummer.2018 sitepc.cc:seasonsummer.2018
## 0.167167 -0.050667
## sitehs.fp:seasonsummer.2019 sitend.eos:seasonsummer.2019
## 0.361667 0.221833
## sitepc.cc:seasonsummer.2019 sitehs.fp:seasonwinter.2018
## -0.029333 0.056333
## sitend.eos:seasonwinter.2018 sitepc.cc:seasonwinter.2018
## 0.003500 0.017667summary (lm2.1)##
## Call:
## lm(formula = cover ~ site + season + site:season, data = all)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.105000 -0.026667 -0.004333 0.018333 0.137667
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.538000 0.039178 13.732 4.24e-16 ***
## sitehs.fp -0.251667 0.055407 -4.542 5.75e-05 ***
## sitend.eos -0.172500 0.061946 -2.785 0.0084 **
## sitepc.cc 0.044000 0.078357 0.562 0.5778
## seasonfall.2019 -0.397000 0.078357 -5.067 1.15e-05 ***
## seasonspring.2019 -0.083333 0.055407 -1.504 0.1411
## seasonsummer.2018 -0.135000 0.055407 -2.437 0.0198 *
## seasonsummer.2019 -0.408667 0.055407 -7.376 9.00e-09 ***
## seasonwinter.2018 -0.088333 0.055407 -1.594 0.1194
## sitehs.fp:seasonfall.2019 0.268667 0.110813 2.424 0.0203 *
## sitend.eos:seasonfall.2019 0.201500 0.114224 1.764 0.0860 .
## sitepc.cc:seasonfall.2019 -0.020000 0.123893 -0.161 0.8726
## sitehs.fp:seasonspring.2019 0.197000 0.078357 2.514 0.0164 *
## sitend.eos:seasonspring.2019 0.004833 0.083110 0.058 0.9539
## sitepc.cc:seasonspring.2019 -0.112333 0.095967 -1.171 0.2493
## sitehs.fp:seasonsummer.2018 0.059333 0.078357 0.757 0.4537
## sitend.eos:seasonsummer.2018 0.167167 0.083110 2.011 0.0516 .
## sitepc.cc:seasonsummer.2018 -0.050667 0.095967 -0.528 0.6007
## sitehs.fp:seasonsummer.2019 0.361667 0.078357 4.616 4.59e-05 ***
## sitend.eos:seasonsummer.2019 0.221833 0.083110 2.669 0.0112 *
## sitepc.cc:seasonsummer.2019 -0.029333 0.095967 -0.306 0.7616
## sitehs.fp:seasonwinter.2018 0.056333 0.078357 0.719 0.4767
## sitend.eos:seasonwinter.2018 0.003500 0.083110 0.042 0.9666
## sitepc.cc:seasonwinter.2018 0.017667 0.095967 0.184 0.8549
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.06786 on 37 degrees of freedom
## Multiple R-squared: 0.8498, Adjusted R-squared: 0.7564
## F-statistic: 9.1 on 23 and 37 DF, p-value: 3.148e-09anova (lm2.1)## Analysis of Variance Table
##
## Response: cover
## Df Sum Sq Mean Sq F value Pr(>F)
## site 3 0.13228 0.044095 9.5758 8.168e-05 ***
## season 5 0.55669 0.111338 24.1784 9.952e-11 ***
## site:season 15 0.27486 0.018324 3.9793 0.0003102 ***
## Residuals 37 0.17038 0.004605
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1plot (lm2.1)## Warning: not plotting observations with leverage one:
## 4, 14, 29, 45, 61
All significant, more significance than previous model, plots look good except again I’m not sure about the residual vs leverage one
####2.2 Is there a difference in Fucus total density between site and time?####
density ~ site as random factor + season as fixed factor
lm2.2 <- lm(no.fuc.q ~ site + factor(season), data =all)
lm2.2##
## Call:
## lm(formula = no.fuc.q ~ site + factor(season), data = all)
##
## Coefficients:
## (Intercept) sitehs.fp
## 64.533 -30.363
## sitend.eos sitepc.cc
## -8.564 -2.281
## factor(season)fall.2019 factor(season)spring.2019
## -30.356 -31.214
## factor(season)summer.2018 factor(season)summer.2019
## 20.411 -31.906
## factor(season)winter.2018
## -32.448summary (lm2.2)##
## Call:
## lm(formula = no.fuc.q ~ site + factor(season), data = all)
##
## Residuals:
## Min 1Q Median 3Q Max
## -52.663 -15.219 -4.604 13.946 80.837
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 64.533 10.053 6.419 4.16e-08 ***
## sitehs.fp -30.363 9.174 -3.310 0.00170 **
## sitend.eos -8.564 9.340 -0.917 0.36338
## sitepc.cc -2.281 9.559 -0.239 0.81234
## factor(season)fall.2019 -30.356 15.643 -1.941 0.05775 .
## factor(season)spring.2019 -31.214 11.511 -2.712 0.00905 **
## factor(season)summer.2018 20.411 11.511 1.773 0.08205 .
## factor(season)summer.2019 -31.906 11.511 -2.772 0.00772 **
## factor(season)winter.2018 -32.448 11.511 -2.819 0.00680 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 25.95 on 52 degrees of freedom
## Multiple R-squared: 0.5107, Adjusted R-squared: 0.4354
## F-statistic: 6.784 on 8 and 52 DF, p-value: 4.606e-06anova (lm2.2)## Analysis of Variance Table
##
## Response: no.fuc.q
## Df Sum Sq Mean Sq F value Pr(>F)
## site 3 8534 2844.5 4.2247 0.009534 **
## factor(season) 5 28008 5601.6 8.3195 7.835e-06 ***
## Residuals 52 35012 673.3
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1plot (lm2.2)
Significant, plots look good, not sure about residual vs leverage TO DO: look into other fits
site as random factor + season as fixed factor + interaction term
lm2.2 <- lm(no.fuc.q ~ site + factor(season) + site:season, data =all)
lm2.2##
## Call:
## lm(formula = no.fuc.q ~ site + factor(season) + site:season,
## data = all)
##
## Coefficients:
## (Intercept) sitehs.fp
## 61.833 -43.767
## sitend.eos sitepc.cc
## -6.933 58.967
## factor(season)fall.2019 factor(season)spring.2019
## -105.700 -103.000
## factor(season)summer.2018 factor(season)summer.2019
## -26.233 -102.800
## factor(season)winter.2018 siteby.rb:seasonfall.2019
## -91.233 94.267
## sitehs.fp:seasonfall.2019 sitend.eos:seasonfall.2019
## 99.833 68.600
## sitepc.cc:seasonfall.2019 siteby.rb:seasonspring.2019
## NA 67.600
## sitehs.fp:seasonspring.2019 sitend.eos:seasonspring.2019
## 106.567 74.300
## sitepc.cc:seasonspring.2019 siteby.rb:seasonsummer.2018
## NA 58.533
## sitehs.fp:seasonsummer.2018 sitend.eos:seasonsummer.2018
## 31.700 57.667
## sitepc.cc:seasonsummer.2018 siteby.rb:seasonsummer.2019
## NA 79.533
## sitehs.fp:seasonsummer.2019 sitend.eos:seasonsummer.2019
## 97.767 67.600
## sitepc.cc:seasonsummer.2019 siteby.rb:seasonwinter.2018
## NA 50.533
## sitehs.fp:seasonwinter.2018 sitend.eos:seasonwinter.2018
## 89.800 56.133
## sitepc.cc:seasonwinter.2018
## NAsummary (lm2.2)##
## Call:
## lm(formula = no.fuc.q ~ site + factor(season) + site:season,
## data = all)
##
## Residuals:
## Min 1Q Median 3Q Max
## -64.567 -4.367 0.000 5.267 68.933
##
## Coefficients: (5 not defined because of singularities)
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 61.833 14.231 4.345 0.000104 ***
## sitehs.fp -43.767 20.126 -2.175 0.036125 *
## sitend.eos -6.933 22.502 -0.308 0.759714
## sitepc.cc 58.967 28.462 2.072 0.045316 *
## factor(season)fall.2019 -105.700 34.859 -3.032 0.004418 **
## factor(season)spring.2019 -103.000 28.462 -3.619 0.000880 ***
## factor(season)summer.2018 -26.233 28.462 -0.922 0.362666
## factor(season)summer.2019 -102.800 28.462 -3.612 0.000897 ***
## factor(season)winter.2018 -91.233 28.462 -3.205 0.002778 **
## siteby.rb:seasonfall.2019 94.267 45.003 2.095 0.043106 *
## sitehs.fp:seasonfall.2019 99.833 45.003 2.218 0.032748 *
## sitend.eos:seasonfall.2019 68.600 46.114 1.488 0.145329
## sitepc.cc:seasonfall.2019 NA NA NA NA
## siteby.rb:seasonspring.2019 67.600 34.859 1.939 0.060128 .
## sitehs.fp:seasonspring.2019 106.567 34.859 3.057 0.004136 **
## sitend.eos:seasonspring.2019 74.300 36.283 2.048 0.047726 *
## sitepc.cc:seasonspring.2019 NA NA NA NA
## siteby.rb:seasonsummer.2018 58.533 34.859 1.679 0.101553
## sitehs.fp:seasonsummer.2018 31.700 34.859 0.909 0.369040
## sitend.eos:seasonsummer.2018 57.667 36.283 1.589 0.120486
## sitepc.cc:seasonsummer.2018 NA NA NA NA
## siteby.rb:seasonsummer.2019 79.533 34.859 2.282 0.028365 *
## sitehs.fp:seasonsummer.2019 97.767 34.859 2.805 0.007980 **
## sitend.eos:seasonsummer.2019 67.600 36.283 1.863 0.070394 .
## sitepc.cc:seasonsummer.2019 NA NA NA NA
## siteby.rb:seasonwinter.2018 50.533 34.859 1.450 0.155583
## sitehs.fp:seasonwinter.2018 89.800 34.859 2.576 0.014120 *
## sitend.eos:seasonwinter.2018 56.133 36.283 1.547 0.130348
## sitepc.cc:seasonwinter.2018 NA NA NA NA
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 24.65 on 37 degrees of freedom
## Multiple R-squared: 0.6858, Adjusted R-squared: 0.4905
## F-statistic: 3.512 on 23 and 37 DF, p-value: 0.0003338anova (lm2.2)## Analysis of Variance Table
##
## Response: no.fuc.q
## Df Sum Sq Mean Sq F value Pr(>F)
## site 3 8533.5 2844.5 4.6817 0.007177 **
## factor(season) 5 28008.2 5601.6 9.2195 9.154e-06 ***
## site:season 15 12531.6 835.4 1.3750 0.210492
## Residuals 37 22480.6 607.6
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1plot (lm2.2)## Warning: not plotting observations with leverage one:
## 4, 14, 29, 45, 61
Significant, plots look good
####REPRODUCTIVE PHENOLOGY (with seasonality)####
3. Is reproduction affected by the season?
####3.1 Is oogonia/receptacle affected by the season?####
lm3.1 <- lm(avg.oog ~ season, data =all)
lm3.1##
## Call:
## lm(formula = avg.oog ~ season, data = all)
##
## Coefficients:
## (Intercept) seasonspring.2019 seasonsummer.2018 seasonsummer.2019
## 25.753 -1.633 -0.429 -10.591
## seasonwinter.2018
## 7.506summary (lm3.1)##
## Call:
## lm(formula = avg.oog ~ season, data = all)
##
## Residuals:
## Min 1Q Median 3Q Max
## -24.120 -11.435 -5.148 11.519 37.454
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 25.753 5.341 4.822 1.28e-05 ***
## seasonspring.2019 -1.633 7.065 -0.231 0.818
## seasonsummer.2018 -0.429 7.065 -0.061 0.952
## seasonsummer.2019 -10.591 7.065 -1.499 0.140
## seasonwinter.2018 7.506 7.065 1.062 0.293
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 16.02 on 52 degrees of freedom
## (4 observations deleted due to missingness)
## Multiple R-squared: 0.1297, Adjusted R-squared: 0.06276
## F-statistic: 1.938 on 4 and 52 DF, p-value: 0.118anova (lm3.1)## Analysis of Variance Table
##
## Response: avg.oog
## Df Sum Sq Mean Sq F value Pr(>F)
## season 4 1989.4 497.36 1.9375 0.118
## Residuals 52 13348.4 256.70plot (lm3.1)
####3.2 Is oogonia/thalli affected by the season?####
lm3.2 <- lm(oog.thalli ~ season, data =all)
lm3.2##
## Call:
## lm(formula = oog.thalli ~ season, data = all)
##
## Coefficients:
## (Intercept) seasonspring.2019 seasonsummer.2018 seasonsummer.2019
## 397819 -118122 -208708 -252414
## seasonwinter.2018
## -260220summary (lm3.2)##
## Call:
## lm(formula = oog.thalli ~ season, data = all)
##
## Residuals:
## Min 1Q Median 3Q Max
## -386196 -170441 -91554 73943 1757053
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 397819 115588 3.442 0.00115 **
## seasonspring.2019 -118122 152908 -0.773 0.44331
## seasonsummer.2018 -208708 152908 -1.365 0.17815
## seasonsummer.2019 -252414 152908 -1.651 0.10482
## seasonwinter.2018 -260220 152908 -1.702 0.09476 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 346800 on 52 degrees of freedom
## (4 observations deleted due to missingness)
## Multiple R-squared: 0.07225, Adjusted R-squared: 0.0008898
## F-statistic: 1.012 on 4 and 52 DF, p-value: 0.4096anova (lm3.2)## Analysis of Variance Table
##
## Response: oog.thalli
## Df Sum Sq Mean Sq F value Pr(>F)
## season 4 4.8698e+11 1.2174e+11 1.0125 0.4096
## Residuals 52 6.2527e+12 1.2024e+11plot (lm3.2)
####3.3 Is conceptacle/thalli affected by the season?####
lm3.3 <- lm(con.thalli ~ season, data =all)
lm3.3##
## Call:
## lm(formula = con.thalli ~ season, data = all)
##
## Coefficients:
## (Intercept) seasonspring.2019 seasonsummer.2018 seasonsummer.2019
## 18468 -3002 -7925 -3119
## seasonwinter.2018
## -10864summary (lm3.3)##
## Call:
## lm(formula = con.thalli ~ season, data = all)
##
## Residuals:
## Min 1Q Median 3Q Max
## -16096 -7231 -2131 5639 57509
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 18468 4245 4.351 6.35e-05 ***
## seasonspring.2019 -3002 5615 -0.535 0.5952
## seasonsummer.2018 -7925 5615 -1.411 0.1641
## seasonsummer.2019 -3119 5615 -0.556 0.5809
## seasonwinter.2018 -10864 5615 -1.935 0.0585 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 12730 on 52 degrees of freedom
## (4 observations deleted due to missingness)
## Multiple R-squared: 0.08934, Adjusted R-squared: 0.01929
## F-statistic: 1.275 on 4 and 52 DF, p-value: 0.2916anova (lm3.3)## Analysis of Variance Table
##
## Response: con.thalli
## Df Sum Sq Mean Sq F value Pr(>F)
## season 4 827224705 206806176 1.2754 0.2916
## Residuals 52 8431680714 162147706plot (lm3.3)
####3.4 Is percent reproductive tissue affected by the season?####
lm3.4 <- lm(perc.rdw ~ season, data =all)
lm3.4##
## Call:
## lm(formula = perc.rdw ~ season, data = all)
##
## Coefficients:
## (Intercept) seasonspring.2019 seasonsummer.2018 seasonsummer.2019
## 24.367 -8.205 -5.058 1.804
## seasonwinter.2018
## -11.406summary (lm3.4)##
## Call:
## lm(formula = perc.rdw ~ season, data = all)
##
## Residuals:
## Min 1Q Median 3Q Max
## -24.083 -10.392 -1.393 9.394 33.425
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 24.367 4.575 5.326 2.18e-06 ***
## seasonspring.2019 -8.205 6.052 -1.356 0.1810
## seasonsummer.2018 -5.058 6.052 -0.836 0.4072
## seasonsummer.2019 1.804 6.052 0.298 0.7668
## seasonwinter.2018 -11.406 6.052 -1.885 0.0651 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 13.73 on 52 degrees of freedom
## (4 observations deleted due to missingness)
## Multiple R-squared: 0.1246, Adjusted R-squared: 0.05727
## F-statistic: 1.85 on 4 and 52 DF, p-value: 0.1332anova (lm3.4)## Analysis of Variance Table
##
## Response: perc.rdw
## Df Sum Sq Mean Sq F value Pr(>F)
## season 4 1394.4 348.59 1.8504 0.1332
## Residuals 52 9796.1 188.39plot (lm3.4)
####3.5 Is adult density affected by the season?####
lm3.5 <- lm(no.large.fuc.q ~ season, data =all)
lm3.5##
## Call:
## lm(formula = no.large.fuc.q ~ season, data = all)
##
## Coefficients:
## (Intercept) seasonfall.2019 seasonspring.2019 seasonsummer.2018
## 5.967 -3.592 3.083 5.646
## seasonsummer.2019 seasonwinter.2018
## -2.450 1.875summary (lm3.5)##
## Call:
## lm(formula = no.large.fuc.q ~ season, data = all)
##
## Residuals:
## Min 1Q Median 3Q Max
## -8.4125 -1.1750 -0.4125 1.2875 11.2875
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 5.967 1.173 5.085 5.32e-06 ***
## seasonfall.2019 -3.592 2.115 -1.698 0.09564 .
## seasonspring.2019 3.083 1.552 1.986 0.05239 .
## seasonsummer.2018 5.646 1.711 3.300 0.00177 **
## seasonsummer.2019 -2.450 1.552 -1.578 0.12069
## seasonwinter.2018 1.875 1.552 1.208 0.23268
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 3.52 on 51 degrees of freedom
## (4 observations deleted due to missingness)
## Multiple R-squared: 0.4264, Adjusted R-squared: 0.3702
## F-statistic: 7.582 on 5 and 51 DF, p-value: 2.184e-05anova (lm3.5)## Analysis of Variance Table
##
## Response: no.large.fuc.q
## Df Sum Sq Mean Sq F value Pr(>F)
## season 5 469.84 93.968 7.5822 2.184e-05 ***
## Residuals 51 632.05 12.393
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1plot (lm3.5)
####3.6 Is juvenile density affected by the season?####
lm3.6 <- lm(no.small.fuc.q ~ season, data =all)
lm3.6##
## Call:
## lm(formula = no.small.fuc.q ~ season, data = all)
##
## Coefficients:
## (Intercept) seasonfall.2019 seasonspring.2019 seasonsummer.2018
## 46.29 -24.79 -32.32 36.59
## seasonsummer.2019 seasonwinter.2018
## -27.48 -32.35summary (lm3.6)##
## Call:
## lm(formula = no.small.fuc.q ~ season, data = all)
##
## Residuals:
## Min 1Q Median 3Q Max
## -68.575 -9.500 -1.667 4.358 70.211
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 46.289 8.521 5.433 1.56e-06 ***
## seasonfall.2019 -24.789 15.361 -1.614 0.11275
## seasonspring.2019 -32.322 11.272 -2.868 0.00600 **
## seasonsummer.2018 36.586 12.421 2.946 0.00485 **
## seasonsummer.2019 -27.481 11.272 -2.438 0.01829 *
## seasonwinter.2018 -32.347 11.272 -2.870 0.00596 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 25.56 on 51 degrees of freedom
## (4 observations deleted due to missingness)
## Multiple R-squared: 0.4954, Adjusted R-squared: 0.446
## F-statistic: 10.02 on 5 and 51 DF, p-value: 1.021e-06anova (lm3.6)## Analysis of Variance Table
##
## Response: no.small.fuc.q
## Df Sum Sq Mean Sq F value Pr(>F)
## season 5 32721 6544.2 10.015 1.021e-06 ***
## Residuals 51 33324 653.4
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1plot (lm3.6)
####REPRODUCTIVE PHENOLOGY (without seasonality)###
4. Is reproductive phenology affected by environmental conditions (salinity, pH, water temperature and air temperature)?
TO DO: check which metric I should be using for each of these variables. For now just sticking with the mean between survey dates but I think mean daily minimun or maximum might be more interesting and insightful
####4.1 Is oogoina/receptical affected by environmental conditions (salinity, pH, water temperature and air temperature)?####
lm4.1 <- lm(avg.oog ~ salinity + ph + water.temp, data =all)
lm4.1##
## Call:
## lm(formula = avg.oog ~ salinity + ph + water.temp, data = all)
##
## Coefficients:
## (Intercept) salinity ph water.temp
## 288.777 1.152 -29.813 -3.675summary (lm4.1)##
## Call:
## lm(formula = avg.oog ~ salinity + ph + water.temp, data = all)
##
## Residuals:
## Min 1Q Median 3Q Max
## -17.635 -9.566 -1.689 6.475 34.053
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 288.7770 125.5277 2.301 0.02700 *
## salinity 1.1523 0.3257 3.537 0.00108 **
## ph -29.8126 15.7331 -1.895 0.06573 .
## water.temp -3.6748 0.8275 -4.441 7.48e-05 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 13.45 on 38 degrees of freedom
## (19 observations deleted due to missingness)
## Multiple R-squared: 0.4057, Adjusted R-squared: 0.3588
## F-statistic: 8.647 on 3 and 38 DF, p-value: 0.000168anova (lm4.1)## Analysis of Variance Table
##
## Response: avg.oog
## Df Sum Sq Mean Sq F value Pr(>F)
## salinity 1 553.1 553.1 3.0584 0.08840 .
## ph 1 571.6 571.6 3.1611 0.08342 .
## water.temp 1 3566.6 3566.6 19.7230 7.477e-05 ***
## Residuals 38 6871.7 180.8
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1plot (lm4.1)
####4.2 Is oogonia/thalli affected by environmental conditions (salinity, pH, water temperature and air temperature)?####
lm4.2 <- lm(oog.thalli ~ salinity + ph + water.temp, data =all)
lm4.2##
## Call:
## lm(formula = oog.thalli ~ salinity + ph + water.temp, data = all)
##
## Coefficients:
## (Intercept) salinity ph water.temp
## 3068710 9392 -375880 -11561summary (lm4.2)##
## Call:
## lm(formula = oog.thalli ~ salinity + ph + water.temp, data = all)
##
## Residuals:
## Min 1Q Median 3Q Max
## -208605 -125253 -24487 61299 855759
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3068710 1784692 1.719 0.0937 .
## salinity 9392 4631 2.028 0.0496 *
## ph -375881 223686 -1.680 0.1011
## water.temp -11561 11764 -0.983 0.3320
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 191200 on 38 degrees of freedom
## (19 observations deleted due to missingness)
## Multiple R-squared: 0.1627, Adjusted R-squared: 0.09662
## F-statistic: 2.462 on 3 and 38 DF, p-value: 0.07734anova (lm4.2)## Analysis of Variance Table
##
## Response: oog.thalli
## Df Sum Sq Mean Sq F value Pr(>F)
## salinity 1 1.3454e+11 1.3454e+11 3.6806 0.06259 .
## ph 1 1.0012e+11 1.0012e+11 2.7389 0.10617
## water.temp 1 3.5297e+10 3.5297e+10 0.9656 0.33199
## Residuals 38 1.3890e+12 3.6553e+10
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1plot (lm4.2)
####4.3 Is conceptacle/thalli affected by environmental conditions (salinity, pH, water temperature and air temperature)?####
lm4.3 <- lm(con.thalli ~ salinity + ph + water.temp, data =all)
lm4.3##
## Call:
## lm(formula = con.thalli ~ salinity + ph + water.temp, data = all)
##
## Coefficients:
## (Intercept) salinity ph water.temp
## 58775.2 161.4 -6342.5 -129.4summary (lm4.3)##
## Call:
## lm(formula = con.thalli ~ salinity + ph + water.temp, data = all)
##
## Residuals:
## Min 1Q Median 3Q Max
## -10380 -5529 -2928 2812 31992
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 58775.2 85416.1 0.688 0.496
## salinity 161.4 221.7 0.728 0.471
## ph -6342.5 10705.7 -0.592 0.557
## water.temp -129.4 563.1 -0.230 0.819
##
## Residual standard error: 9150 on 38 degrees of freedom
## (19 observations deleted due to missingness)
## Multiple R-squared: 0.02485, Adjusted R-squared: -0.05213
## F-statistic: 0.3229 on 3 and 38 DF, p-value: 0.8088anova (lm4.3)## Analysis of Variance Table
##
## Response: con.thalli
## Df Sum Sq Mean Sq F value Pr(>F)
## salinity 1 47864762 47864762 0.5717 0.4543
## ph 1 28807975 28807975 0.3441 0.5610
## water.temp 1 4423765 4423765 0.0528 0.8194
## Residuals 38 3181730439 83729748plot (lm4.3)
####4.4 Is percent reproductive tissue affected by environmental conditions (salinity, pH, water temperature and air temperature)?####
lm4.4 <- lm(perc.rdw ~ salinity + ph + water.temp, data =all)
lm4.4##
## Call:
## lm(formula = perc.rdw ~ salinity + ph + water.temp, data = all)
##
## Coefficients:
## (Intercept) salinity ph water.temp
## 329.7386 0.2022 -40.9228 0.3886summary (lm4.4)##
## Call:
## lm(formula = perc.rdw ~ salinity + ph + water.temp, data = all)
##
## Residuals:
## Min 1Q Median 3Q Max
## -17.097 -11.126 -1.268 9.220 29.659
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 329.7386 122.9021 2.683 0.0107 *
## salinity 0.2022 0.3189 0.634 0.5298
## ph -40.9228 15.4040 -2.657 0.0115 *
## water.temp 0.3886 0.8102 0.480 0.6343
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 13.17 on 38 degrees of freedom
## (19 observations deleted due to missingness)
## Multiple R-squared: 0.1886, Adjusted R-squared: 0.1246
## F-statistic: 2.945 on 3 and 38 DF, p-value: 0.04509anova (lm4.4)## Analysis of Variance Table
##
## Response: perc.rdw
## Df Sum Sq Mean Sq F value Pr(>F)
## salinity 1 255.6 255.57 1.4743 0.23216
## ph 1 1236.0 1236.03 7.1303 0.01109 *
## water.temp 1 39.9 39.87 0.2300 0.63426
## Residuals 38 6587.2 173.35
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1plot (lm4.4)
####4.5 Is juvenile density affected by environmental conditions (salinity, pH, water temperature and air temperature)?####
lm4.5 <- lm(no.small.fuc.q ~ salinity + ph + water.temp, data =all)
lm4.5##
## Call:
## lm(formula = no.small.fuc.q ~ salinity + ph + water.temp, data = all)
##
## Coefficients:
## (Intercept) salinity ph water.temp
## -196.608 1.501 14.907 4.941summary (lm4.5)##
## Call:
## lm(formula = no.small.fuc.q ~ salinity + ph + water.temp, data = all)
##
## Residuals:
## Min 1Q Median 3Q Max
## -48.774 -20.807 -6.178 19.982 83.552
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -196.6076 314.3187 -0.626 0.5354
## salinity 1.5011 0.7607 1.973 0.0558 .
## ph 14.9069 39.4949 0.377 0.7079
## water.temp 4.9405 1.8316 2.697 0.0104 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 31.22 on 38 degrees of freedom
## (19 observations deleted due to missingness)
## Multiple R-squared: 0.3624, Adjusted R-squared: 0.3121
## F-statistic: 7.2 on 3 and 38 DF, p-value: 0.0006082anova (lm4.5)## Analysis of Variance Table
##
## Response: no.small.fuc.q
## Df Sum Sq Mean Sq F value Pr(>F)
## salinity 1 13787 13787.0 14.1477 0.0005693 ***
## ph 1 172 172.2 0.1767 0.6766207
## water.temp 1 7091 7090.6 7.2761 0.0103617 *
## Residuals 38 37031 974.5
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1plot (lm4.5)
####4.6 Is adult density affected by environmental conditions (salinity, pH, water temperature and air temperature)?####
lm4.6 <- lm(no.large.fuc.q ~ salinity + ph + water.temp, data =all)
lm4.6##
## Call:
## lm(formula = no.large.fuc.q ~ salinity + ph + water.temp, data = all)
##
## Coefficients:
## (Intercept) salinity ph water.temp
## -11.428619 0.009459 3.089848 -0.368876summary (lm4.6)##
## Call:
## lm(formula = no.large.fuc.q ~ salinity + ph + water.temp, data = all)
##
## Residuals:
## Min 1Q Median 3Q Max
## -6.057 -3.407 -1.521 3.009 16.395
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -11.428619 49.474480 -0.231 0.819
## salinity 0.009459 0.119731 0.079 0.937
## ph 3.089848 6.216584 0.497 0.622
## water.temp -0.368876 0.288292 -1.280 0.208
##
## Residual standard error: 4.914 on 38 degrees of freedom
## (19 observations deleted due to missingness)
## Multiple R-squared: 0.05791, Adjusted R-squared: -0.01646
## F-statistic: 0.7786 on 3 and 38 DF, p-value: 0.5132anova (lm4.6)## Analysis of Variance Table
##
## Response: no.large.fuc.q
## Df Sum Sq Mean Sq F value Pr(>F)
## salinity 1 11.38 11.382 0.4714 0.4965
## ph 1 5.49 5.488 0.2273 0.6362
## water.temp 1 39.53 39.528 1.6372 0.2085
## Residuals 38 917.47 24.144plot (lm4.6)
