Birds_Planted_Forest_analysis_by_plot
Ron Chen
2024-01-31
## Loading required package: readxl## Loading required package: lubridate##
## Attaching package: 'lubridate'## The following objects are masked from 'package:data.table':
##
## hour, isoweek, mday, minute, month, quarter, second, wday, week,
## yday, year## The following objects are masked from 'package:base':
##
## date, intersect, setdiff, union## Loading required package: solartime## - Winter data excluded.
## - The counts of all passing / wintering species (not breeding in the sampled unit) were set to zero. Abreed_and_non_breed contains the non-breeders as well.
## - Species less likely to be interacting with sampling location were EXCLUDED
## - observations with counts greater than or equal to 10 standard deviations AND greater than or equal to 60 individuals were set to 1,
## under the assumption that these are migrating or local wandering phenomena.
## No outliers found in unit Arid South.
## 2 outliers set equal to 1 in Herbaceous and Dwarf-Shrub Vegetation unit.
## unit point_name year
## 1: Herbaceous and Dwarf-Shrub Vegetation Gamla Near 3 2014
## 2: Herbaceous and Dwarf-Shrub Vegetation Yavneel Far 2 2018
## SciName count_under_250 Z_score
## 1: Passer domesticus 62 11.90316
## 2: Passer hispaniolensis 60 11.35733
## No outliers found in unit Inland Sands.
## No outliers found in unit Loess Covered Areas in the Northern Negev.
## No outliers found in unit Mediterranean-Desert Transition Zone.
## 5 outliers set equal to 1 in Mediterranean Maquis unit.
## unit point_name year SciName
## 1: Mediterranean Maquis Beit Oren Near 1 2019 Curruca curruca
## 2: Mediterranean Maquis Beit Oren Near 3 2019 Curruca curruca
## 3: Mediterranean Maquis Kerem Maharal Near 3 2021 Chloris chloris
## 4: Mediterranean Maquis Kerem Maharal Near 3 2021 Garrulus glandarius
## 5: Mediterranean Maquis Nir Etzion Far 11 2021 Columba livia
## count_under_250 Z_score
## 1: 99 11.94535
## 2: 142 17.16927
## 3: 80 19.19266
## 4: 150 20.50138
## 5: 100 14.85118
## 2 outliers set equal to 1 in Negev Highlands unit.
## unit point_name year SciName
## 1: Negev Highlands Yeruham Far Slope 5 2018 Carpospiza brachydactyla
## 2: Negev Highlands Bislach Near Slope 6 2020 Passer hispaniolensis
## count_under_250 Z_score
## 1: 70 14.34802
## 2: 100 10.61343
## No outliers found in unit Planted Conifer Forest.Planted Conifer Forest
Unit is divided into 3 subunits: Judea, Carmel and Galilee. Only factors is time. All plots are located far from settlements. Sampling started in spring 2014. Total 5 campaigns, 3 subunits per campaign, 5 sites per subunit, with 3 plots per site. Total of 45 plots per campaign, and 225 plot-campaign combinations.
Raw data Total abundance: 6230 Number of observations: 2161 Total richness: 86
Filtered data Total abundance: 4150 Number of observations: 1355 Total richness: 43
Model gma, abundance and richness
Richness done with rare species. Abundance and mean abundance done without rare species. Full models include cosine and sine of the time difference from June 21st (in radians).
richness
Explore data.
## [1] "RICHNESS WITH RARE SPECIES"
| richness | year_ct | site | subunit | td_sc | cos_td_rad | sin_td_rad | h_from_sunrise | cos_hsun | sin_hsun | monitors_name | wind | precipitation | temperature | clouds | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Min. : 1.000 | Min. :2.0 | Aderet : 15 | Judean Highlands:75 | Min. :-2.3697 | Min. :-0.3335 | Min. :-0.9959 | Length:225 | Min. :-0.2639 | Min. :-0.1201 | Eyal Shochat: 35 | 0 : 24 | 0 : 56 | 0 : 0 | 0 : 9 | |
| 1st Qu.: 5.000 | 1st Qu.:3.0 | Amatzia : 15 | Carmel :75 | 1st Qu.:-0.2713 | 1st Qu.: 0.5702 | 1st Qu.:-0.8215 | Class :difftime | 1st Qu.: 0.6155 | 1st Qu.: 0.2784 | Asaf Mayrose: 23 | 1 : 13 | 3 : 3 | 1 : 20 | 1 : 3 | |
| Median : 6.000 | Median :5.0 | Bat Shlomo: 15 | Galilee :75 | Median : 0.3010 | Median : 0.7611 | Median :-0.6486 | Mode :numeric | Median : 0.8098 | Median : 0.5867 | Sassi Haham : 18 | 2 : 7 | NA’s:166 | 2 : 26 | 2 : 1 | |
| Mean : 6.022 | Mean :5.2 | Eitanim : 15 | NA | Mean : 0.2127 | Mean : 0.6933 | Mean :-0.6326 | NA | Mean : 0.7332 | Mean : 0.5336 | Eliraz Dvir : 12 | 3 : 1 | NA | 3 : 1 | 3 : 2 | |
| 3rd Qu.: 7.000 | 3rd Qu.:7.0 | Elyakim : 15 | NA | 3rd Qu.: 0.6444 | 3rd Qu.: 0.8521 | 3rd Qu.:-0.5234 | NA | 3rd Qu.: 0.9605 | 3rd Qu.: 0.7881 | Eran Banker : 9 | NA’s:180 | NA | NA’s:178 | NA’s:210 | |
| Max. :13.000 | Max. :9.0 | Eshtaol : 15 | NA | Max. : 1.6364 | Max. : 0.9947 | Max. :-0.1031 | NA | Max. : 1.0000 | Max. : 1.0000 | (Other) : 20 | NA | NA | NA | NA | |
| NA | NA | (Other) :135 | NA | NA | NA | NA | NA | NA’s :45 | NA’s :45 | NA’s :108 | NA | NA | NA | NA |
no observation for all 4 weather variables. many NAs for sampling time of day variables.exclude from model. An extreme observation of richness>20 in Judean highlands near settlements.
| richness | year_ct | site | subunit | td_sc | cos_td_rad | sin_td_rad | h_from_sunrise | cos_hsun | sin_hsun | |
|---|---|---|---|---|---|---|---|---|---|---|
| richness | 1.0000000 | -0.0869667 | -0.1526824 | -0.2098492 | 0.2468122 | 0.2620866 | 0.2032932 | -0.1291245 | 0.1729942 | -0.0748826 |
| year_ct | -0.0869667 | 1.0000000 | 0.0000000 | 0.0000000 | -0.4304647 | -0.4084156 | -0.4336689 | -0.0881196 | 0.0985736 | -0.0618585 |
| site | -0.1526824 | 0.0000000 | 1.0000000 | 0.7181325 | -0.1237969 | -0.0708077 | -0.2124888 | 0.2856522 | -0.2977022 | 0.2576585 |
| subunit | -0.2098492 | 0.0000000 | 0.7181325 | 1.0000000 | -0.1180592 | -0.0606624 | -0.2128418 | 0.2944203 | -0.3327003 | 0.2303051 |
| td_sc | 0.2468122 | -0.4304647 | -0.1237969 | -0.1180592 | 1.0000000 | 0.9753453 | 0.9307073 | 0.0619157 | -0.0348851 | 0.0841439 |
| cos_td_rad | 0.2620866 | -0.4084156 | -0.0708077 | -0.0606624 | 0.9753453 | 1.0000000 | 0.8324627 | 0.1128374 | -0.0873088 | 0.1309387 |
| sin_td_rad | 0.2032932 | -0.4336689 | -0.2124888 | -0.2128418 | 0.9307073 | 0.8324627 | 1.0000000 | -0.0274966 | 0.0538205 | -0.0012408 |
| h_from_sunrise | -0.1291245 | -0.0881196 | 0.2856522 | 0.2944203 | 0.0619157 | 0.1128374 | -0.0274966 | 1.0000000 | -0.9603558 | 0.9618286 |
| cos_hsun | 0.1729942 | 0.0985736 | -0.2977022 | -0.3327003 | -0.0348851 | -0.0873088 | 0.0538205 | -0.9603558 | 1.0000000 | -0.8509778 |
| sin_hsun | -0.0748826 | -0.0618585 | 0.2576585 | 0.2303051 | 0.0841439 | 0.1309387 | -0.0012408 | 0.9618286 | -0.8509778 | 1.0000000 |
Fit Poisson glm, check for existence of overdispersion
## [1] "Estimating overdispersion parameter phi: (Res. Dev.)/(n-p) where n=number of observations; p=number of parameters in the model."## [1] "od = 0.658637694737327"Overdispersion parameter is < 1. Choose Poisson.
## Warning in checkConv(attr(opt, "derivs"), opt$par, ctrl = control$checkConv, :
## Model failed to converge with max|grad| = 0.00332887 (tol = 0.002, component 1)go with mixed model with subunits as a fixed effect (want to make statement about subunits), attempt model selection yet.
## Generalized linear mixed model fit by maximum likelihood (Laplace
## Approximation) [glmerMod]
## Family: poisson ( log )
## Formula: richness ~ subunit * year_ct + cos_td_rad + sin_td_rad + (1 |
## site)
## Data: P.anal
##
## AIC BIC logLik deviance df.resid
## 965.2 995.9 -473.6 947.2 216
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -1.7752 -0.5322 -0.0193 0.4039 2.7190
##
## Random effects:
## Groups Name Variance Std.Dev.
## site (Intercept) 0.01239 0.1113
## Number of obs: 225, groups: site, 15
##
## Fixed effects:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 1.04905 0.33058 3.173 0.00151 **
## subunitCarmel 0.13102 0.17236 0.760 0.44718
## subunitGalilee -0.07910 0.17679 -0.447 0.65459
## year_ct 0.01929 0.02006 0.961 0.33638
## cos_td_rad 0.67536 0.24778 2.726 0.00642 **
## sin_td_rad -0.39325 0.28670 -1.372 0.17017
## subunitCarmel:year_ct -0.03749 0.02601 -1.441 0.14950
## subunitGalilee:year_ct -0.02178 0.02704 -0.805 0.42054
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) sbntCr sbntGl yer_ct cs_td_ sn_td_ sbnC:_
## subunitCrml -0.029
## subunitGall -0.026 0.548
## year_ct -0.148 0.622 0.603
## cos_td_rad -0.933 -0.180 -0.179 -0.050
## sin_td_rad 0.841 0.317 0.307 0.300 -0.844
## sbntCrml:y_ 0.131 -0.808 -0.426 -0.708 0.034 -0.158
## sbntGll:yr_ 0.045 -0.439 -0.823 -0.688 0.120 -0.227 0.499
## optimizer (Nelder_Mead) convergence code: 0 (OK)
## Model failed to converge with max|grad| = 0.00332887 (tol = 0.002, component 1)perform stepwise model selection of poisson mixed model.
## Single term deletions
##
## Model:
## richness ~ subunit * year_ct + cos_td_rad + sin_td_rad + (1 |
## site)
## npar AIC
## <none> 965.18
## cos_td_rad 1 971.36
## sin_td_rad 1 965.10
## subunit:year_ct 2 963.27remove subunit X year.
## Single term deletions
##
## Model:
## richness ~ subunit + year_ct + cos_td_rad + sin_td_rad + (1 |
## site)
## npar AIC
## <none> 963.27
## subunit 2 962.88
## year_ct 1 961.30
## cos_td_rad 1 969.76
## sin_td_rad 1 964.06drop year.
## Single term deletions
##
## Model:
## richness ~ subunit + cos_td_rad + sin_td_rad + (1 | site)
## npar AIC
## <none> 961.30
## subunit 2 960.89
## cos_td_rad 1 967.83
## sin_td_rad 1 962.09drop subunit.
## Single term deletions
##
## Model:
## richness ~ cos_td_rad + sin_td_rad + (1 | site)
## npar AIC
## <none> 960.89
## cos_td_rad 1 967.34
## sin_td_rad 1 961.41drop sine.
## Single term deletions
##
## Model:
## richness ~ cos_td_rad + (1 | site)
## npar AIC
## <none> 961.41
## cos_td_rad 1 968.85only time of year remains. Final model:
## Generalized linear mixed model fit by maximum likelihood (Laplace
## Approximation) [glmerMod]
## Family: poisson ( log )
## Formula: richness ~ cos_td_rad + (1 | site)
## Data: P.anal
##
## AIC BIC logLik deviance df.resid
## 961.4 971.7 -477.7 955.4 222
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -1.81116 -0.55209 -0.05948 0.41685 2.46546
##
## Random effects:
## Groups Name Variance Std.Dev.
## site (Intercept) 0.01481 0.1217
## Number of obs: 225, groups: site, 15
##
## Fixed effects:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 1.54330 0.09217 16.745 < 2e-16 ***
## cos_td_rad 0.34666 0.11534 3.005 0.00265 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr)
## cos_td_rad -0.892## $site
## (Intercept)
## Aderet 0.041958260
## Amatzia 0.194200215
## Bat Shlomo 0.050085708
## Eitanim -0.057342406
## Elyakim 0.017471976
## Eshtaol -0.142787807
## Givat Yeshayahu 0.134084761
## Kabri -0.117575421
## Kerem Maharal 0.066797701
## Manara -0.094902416
## Meron -0.110849885
## Ofer -0.009267953
## Ramat Hashofet 0.008894432
## Ramot Naftali 0.055138130
## Zuriel -0.008776173
##
## with conditional variances for "site"## $site
attempt to add year, see if significant:
## Generalized linear mixed model fit by maximum likelihood (Laplace
## Approximation) [glmerMod]
## Family: poisson ( log )
## Formula: richness ~ year_ct + cos_td_rad + (1 | site)
## Data: P.anal
##
## AIC BIC logLik deviance df.resid
## 963.4 977.0 -477.7 955.4 221
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -1.80811 -0.54823 -0.05958 0.40392 2.42124
##
## Random effects:
## Groups Name Variance Std.Dev.
## site (Intercept) 0.01476 0.1215
## Number of obs: 225, groups: site, 15
##
## Fixed effects:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 1.522036 0.133873 11.369 < 2e-16 ***
## year_ct 0.002545 0.011607 0.219 0.82644
## cos_td_rad 0.358225 0.126935 2.822 0.00477 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) yer_ct
## year_ct -0.725
## cos_td_rad -0.860 0.416year not significant, rightfully dropped.
No statistically significant effects of subunit or time found for richness.
geometric mean of abundance
Explore data. Exclude time of day because of high number of NAs.
## [1] "GEOMETRIC MEAN ABUNDANCE WITHOUT RARE SPECIES"
| gma | year_ct | site | settlements | subunit | td_sc | cos_td_rad | sin_td_rad | |
|---|---|---|---|---|---|---|---|---|
| Min. :1.000 | Min. :2.0 | Aderet : 15 | Far : 0 | Judean Highlands:75 | Min. :-2.3697 | Min. :-0.3335 | Min. :-0.9959 | |
| 1st Qu.:1.861 | 1st Qu.:3.0 | Amatzia : 15 | Near: 0 | Carmel :75 | 1st Qu.:-0.2713 | 1st Qu.: 0.5702 | 1st Qu.:-0.8215 | |
| Median :2.330 | Median :5.0 | Bat Shlomo: 15 | NA’s:225 | Galilee :75 | Median : 0.3010 | Median : 0.7611 | Median :-0.6486 | |
| Mean :2.574 | Mean :5.2 | Eitanim : 15 | NA | NA | Mean : 0.2127 | Mean : 0.6933 | Mean :-0.6326 | |
| 3rd Qu.:3.061 | 3rd Qu.:7.0 | Elyakim : 15 | NA | NA | 3rd Qu.: 0.6444 | 3rd Qu.: 0.8521 | 3rd Qu.:-0.5234 | |
| Max. :6.327 | Max. :9.0 | Eshtaol : 15 | NA | NA | Max. : 1.6364 | Max. : 0.9947 | Max. :-0.1031 | |
| NA | NA | (Other) :135 | NA | NA | NA | NA | NA |
Fit glm, compare gamma, gaussian (poisson inappropriate because response is not discrete)
Remove rows 67, 112.
Fit fixed and mixed models.
## Warning in checkConv(attr(opt, "derivs"), opt$par, ctrl = control$checkConv, :
## Model failed to converge with max|grad| = 0.0245391 (tol = 0.002, component 1)Mixed model did not converge, attempt model selection.
## Generalized linear mixed model fit by maximum likelihood (Laplace
## Approximation) [glmerMod]
## Family: Gamma ( inverse )
## Formula: gma ~ subunit * year_ct + cos_td_rad + sin_td_rad + (1 | site)
## Data: P.anal
##
## AIC BIC logLik deviance df.resid
## 564.6 598.7 -272.3 544.6 213
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -1.6655 -0.6553 -0.1655 0.5333 3.0425
##
## Random effects:
## Groups Name Variance Std.Dev.
## site (Intercept) 0.0003395 0.01843
## Residual 0.1162229 0.34091
## Number of obs: 223, groups: site, 15
##
## Fixed effects:
## Estimate Std. Error t value Pr(>|z|)
## (Intercept) 0.539774 0.111683 4.833 1.34e-06 ***
## subunitCarmel -0.016579 0.050921 -0.326 0.74474
## subunitGalilee 0.157705 0.059452 2.653 0.00799 **
## year_ct 0.011174 0.006922 1.614 0.10649
## cos_td_rad -0.146400 0.083722 -1.749 0.08035 .
## sin_td_rad 0.159644 0.094525 1.689 0.09124 .
## subunitCarmel:year_ct -0.006663 0.008195 -0.813 0.41616
## subunitGalilee:year_ct -0.016063 0.009618 -1.670 0.09489 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) sbntCr sbntGl yer_ct cs_td_ sn_td_ sbnC:_
## subunitCrml -0.024
## subunitGall -0.013 0.546
## year_ct -0.195 0.676 0.582
## cos_td_rad -0.942 -0.174 -0.157 0.006
## sin_td_rad 0.850 0.329 0.291 0.245 -0.845
## sbntCrml:y_ 0.176 -0.830 -0.450 -0.780 -0.002 -0.135
## sbntGll:yr_ 0.056 -0.469 -0.863 -0.671 0.096 -0.206 0.535
## optimizer (Nelder_Mead) convergence code: 0 (OK)
## Model failed to converge with max|grad| = 0.0245391 (tol = 0.002, component 1)perform stepwise model selection of gaussian model.
## Warning in checkConv(attr(opt, "derivs"), opt$par, ctrl = control$checkConv, :
## Model failed to converge with max|grad| = 0.00783015 (tol = 0.002, component 1)## Warning in checkConv(attr(opt, "derivs"), opt$par, ctrl = control$checkConv, :
## Model failed to converge with max|grad| = 0.00313316 (tol = 0.002, component 1)## Warning in checkConv(attr(opt, "derivs"), opt$par, ctrl = control$checkConv, :
## Model failed to converge with max|grad| = 0.00368728 (tol = 0.002, component 1)## Single term deletions
##
## Model:
## gma ~ subunit * year_ct + cos_td_rad + sin_td_rad + (1 | site)
## npar AIC
## <none> 564.62
## cos_td_rad 1 565.95
## sin_td_rad 1 565.58
## subunit:year_ct 2 563.40drop year : subunit
## Warning in checkConv(attr(opt, "derivs"), opt$par, ctrl = control$checkConv, :
## Model failed to converge with max|grad| = 0.00368728 (tol = 0.002, component 1)## Single term deletions
##
## Model:
## gma ~ subunit + year_ct + cos_td_rad + sin_td_rad + (1 | site)
## npar AIC
## <none> 563.40
## subunit 2 569.61
## year_ct 1 562.38
## cos_td_rad 1 564.21
## sin_td_rad 1 563.39drop year
## Single term deletions
##
## Model:
## gma ~ subunit + cos_td_rad + sin_td_rad + (1 | site)
## npar AIC
## <none> 562.38
## subunit 2 568.55
## cos_td_rad 1 563.38
## sin_td_rad 1 561.91drop sine
## Single term deletions
##
## Model:
## gma ~ subunit + cos_td_rad + (1 | site)
## npar AIC
## <none> 561.91
## subunit 2 569.17
## cos_td_rad 1 561.68drop cosine.
## Single term deletions
##
## Model:
## gma ~ subunit + (1 | site)
## npar AIC
## <none> 561.68
## subunit 2 569.51Only subunit remains. This is the final model:
## Generalized linear mixed model fit by maximum likelihood (Laplace
## Approximation) [glmerMod]
## Family: Gamma ( inverse )
## Formula: gma ~ subunit + (1 | site)
## Data: P.anal
##
## AIC BIC logLik deviance df.resid
## 561.7 578.7 -275.8 551.7 218
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -1.6633 -0.6407 -0.1850 0.5666 3.2713
##
## Random effects:
## Groups Name Variance Std.Dev.
## site (Intercept) 0.0002287 0.01512
## Residual 0.1224116 0.34987
## Number of obs: 223, groups: site, 15
##
## Fixed effects:
## Estimate Std. Error t value Pr(>|z|)
## (Intercept) 0.40298 0.01839 21.915 < 2e-16 ***
## subunitCarmel -0.06942 0.02443 -2.841 0.00449 **
## subunitGalilee 0.06403 0.02772 2.310 0.02088 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) sbntCr
## subunitCrml -0.748
## subunitGall -0.662 0.497## $site
## (Intercept)
## Aderet 0.0014332238
## Amatzia 0.0052545903
## Bat Shlomo 0.0053741289
## Eitanim 0.0011619522
## Elyakim 0.0030863418
## Eshtaol -0.0063275188
## Givat Yeshayahu -0.0019589870
## Kabri -0.0087587560
## Kerem Maharal -0.0092614993
## Manara 0.0043994590
## Meron 0.0011939807
## Ofer -0.0017370411
## Ramat Hashofet 0.0017625159
## Ramot Naftali 0.0027354100
## Zuriel 0.0002541877
##
## with conditional variances for "site"## $site
## Registered S3 methods overwritten by 'broom':
## method from
## tidy.glht jtools
## tidy.summary.glht jtools| Observations | 223 |
| Dependent variable | gma |
| Type | Mixed effects generalized linear model |
| Family | Gamma |
| Link | inverse |
| AIC | 561.68 |
| BIC | 578.72 |
| Pseudo-R² (fixed effects) | NA |
| Pseudo-R² (total) | NA |
| Est. | S.E. | t val. | p | |
|---|---|---|---|---|
| (Intercept) | 0.40 | 0.02 | 21.92 | 0.00 |
| subunitCarmel | -0.07 | 0.02 | -2.84 | 0.00 |
| subunitGalilee | 0.06 | 0.03 | 2.31 | 0.02 |
| Group | Parameter | Std. Dev. |
|---|---|---|
| site | (Intercept) | 0.02 |
| Residual | 0.35 |
| Group | # groups | ICC |
|---|---|---|
| site | 15 | 0.00 |
## Confidence intervals for merMod models is an experimental feature. The
## intervals reflect only the variance of the fixed effects, not the random
## effects.
## Confidence intervals for merMod models is an experimental feature. The
## intervals reflect only the variance of the fixed effects, not the random
## effects.
## subunit emmean SE df asymp.LCL asymp.UCL
## Judean Highlands 0.403 0.0184 Inf 0.367 0.439
## Carmel 0.334 0.0162 Inf 0.302 0.365
## Galilee 0.467 0.0208 Inf 0.426 0.508
##
## Results are given on the inverse (not the response) scale.
## Confidence level used: 0.95## [1] 39.82036## [1] 20.65868## [1] 15.88089| contrast | estimate | SE | df | z.ratio | p.value |
|---|---|---|---|---|---|
| Judean Highlands - Carmel | 0.0694207 | 0.0244337 | Inf | 2.841183 | 0.0067420 |
| Judean Highlands - Galilee | -0.0640337 | 0.0277188 | Inf | -2.310115 | 0.0208818 |
| Carmel - Galilee | -0.1334544 | 0.0263182 | Inf | -5.070793 | 0.0000012 |
GMA is significantly different among all three subunits, with the highest being Carmel and the lowest Galilee (post-hoc z-test for pairwise contrasts, fdr correction for 3 tests). No significant temporal trend.
abundance
Explore data
## [1] "ABUNDANCE WITHOUT RARE SPECIES"
| abundance | year_ct | site | settlements | subunit | td_sc | cos_td_rad | sin_td_rad | |
|---|---|---|---|---|---|---|---|---|
| Min. : 1.00 | Min. :2.0 | Aderet : 15 | Far : 0 | Judean Highlands:75 | Min. :-2.3697 | Min. :-0.3335 | Min. :-0.9959 | |
| 1st Qu.:10.00 | 1st Qu.:3.0 | Amatzia : 15 | Near: 0 | Carmel :75 | 1st Qu.:-0.2713 | 1st Qu.: 0.5702 | 1st Qu.:-0.8215 | |
| Median :16.00 | Median :5.0 | Bat Shlomo: 15 | NA’s:225 | Galilee :75 | Median : 0.3010 | Median : 0.7611 | Median :-0.6486 | |
| Mean :17.14 | Mean :5.2 | Eitanim : 15 | NA | NA | Mean : 0.2127 | Mean : 0.6933 | Mean :-0.6326 | |
| 3rd Qu.:21.00 | 3rd Qu.:7.0 | Elyakim : 15 | NA | NA | 3rd Qu.: 0.6444 | 3rd Qu.: 0.8521 | 3rd Qu.:-0.5234 | |
| Max. :69.00 | Max. :9.0 | Eshtaol : 15 | NA | NA | Max. : 1.6364 | Max. : 0.9947 | Max. :-0.1031 | |
| NA | NA | (Other) :135 | NA | NA | NA | NA | NA |
Two outliers with total abundance >50. Examine:
## subunit point_name datetime monitors_name
## 1: Judean Highlands Aderet KKL Plantings 3 2017-05-25 07:45:00 <NA>
## 2: Carmel Elyakim KKL Plantings 1 2019-04-26 06:07:00 Other
## richness gma abundance
## 1: 11 3.684917 54
## 2: 6 5.614404 69Keep all plots for now.
PHI>1, hence choose negative binomial. Fit fixed and mixed models. Choose mixed model if possible, otherwise choose a model with fixed-effects only.
## Warning in checkConv(attr(opt, "derivs"), opt$par, ctrl = control$checkConv, :
## Model failed to converge with max|grad| = 0.00396002 (tol = 0.002, component 1)mixed model failed to converge, attempt model selection of mixed model and see if eventually it converges.
## Warning in checkConv(attr(opt, "derivs"), opt$par, ctrl = control$checkConv, :
## Model failed to converge with max|grad| = 0.00412153 (tol = 0.002, component 1)## Warning in checkConv(attr(opt, "derivs"), opt$par, ctrl = control$checkConv, :
## Model failed to converge with max|grad| = 0.00794915 (tol = 0.002, component 1)## Single term deletions
##
## Model:
## abundance ~ subunit * year_ct + cos_td_rad + sin_td_rad + (1 |
## site)
## npar AIC
## <none> 1582.7
## cos_td_rad 1 1588.9
## sin_td_rad 1 1584.8
## subunit:year_ct 2 1579.3drop subunit X year
## Single term deletions
##
## Model:
## abundance ~ subunit + year_ct + cos_td_rad + sin_td_rad + (1 |
## site)
## npar AIC
## <none> 1579.3
## subunit 2 1592.1
## year_ct 1 1577.7
## cos_td_rad 1 1585.1
## sin_td_rad 1 1581.2drop year
## Single term deletions
##
## Model:
## abundance ~ subunit + cos_td_rad + sin_td_rad + (1 | site)
## npar AIC
## <none> 1577.7
## subunit 2 1590.6
## cos_td_rad 1 1583.8
## sin_td_rad 1 1579.3drop sine because \(\Delta AIC<2\)
## Single term deletions
##
## Model:
## abundance ~ subunit + cos_td_rad + (1 | site)
## npar AIC
## <none> 1579.2
## subunit 2 1595.5
## cos_td_rad 1 1582.8subunit and time of year remain. The final model:
## Generalized linear mixed model fit by maximum likelihood (Laplace
## Approximation) [glmerMod]
## Family: Negative Binomial(5.0582) ( log )
## Formula: abundance ~ subunit + cos_td_rad + (1 | site)
## Data: P.anal
##
## AIC BIC logLik deviance df.resid
## 1579.2 1599.7 -783.6 1567.2 219
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -1.7006 -0.6340 -0.2512 0.4708 4.8965
##
## Random effects:
## Groups Name Variance Std.Dev.
## site (Intercept) 0.001695 0.04117
## Number of obs: 225, groups: site, 15
##
## Fixed effects:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 2.66798 0.11700 22.803 < 2e-16 ***
## subunitCarmel 0.14408 0.08567 1.682 0.0926 .
## subunitGalilee -0.41092 0.08844 -4.646 3.38e-06 ***
## cos_td_rad 0.33290 0.13893 2.396 0.0166 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) sbntCr sbntGl
## subunitCrml -0.390
## subunitGall -0.399 0.491
## cos_td_rad -0.852 0.021 0.047## $site
## (Intercept)
## Aderet 0.002221368
## Amatzia 0.003288417
## Bat Shlomo -0.009378040
## Eitanim -0.016330075
## Elyakim 0.002662661
## Eshtaol -0.013511033
## Givat Yeshayahu 0.024408095
## Kabri 0.007108884
## Kerem Maharal 0.012060266
## Manara -0.010333935
## Meron -0.018844743
## Ofer 0.006270886
## Ramat Hashofet -0.011544826
## Ramot Naftali 0.015570814
## Zuriel 0.006594637
##
## with conditional variances for "site"## $site
Interpretation of abundance model:
| Observations | 225 |
| Dependent variable | abundance |
| Type | Mixed effects generalized linear model |
| Family | Negative Binomial(5.0582) |
| Link | log |
| AIC | 1579.156 |
| BIC | 1599.653 |
| Pseudo-R² (fixed effects) | 0.524 |
| Pseudo-R² (total) | 0.537 |
| exp(Est.) | S.E. | z val. | p | |
|---|---|---|---|---|
| (Intercept) | 14.411 | 0.117 | 22.803 | 0.000 |
| subunitCarmel | 1.155 | 0.086 | 1.682 | 0.093 |
| subunitGalilee | 0.663 | 0.088 | -4.646 | 0.000 |
| cos_td_rad | 1.395 | 0.139 | 2.396 | 0.017 |
| Group | Parameter | Std. Dev. |
|---|---|---|
| site | (Intercept) | 0.041 |
| Group | # groups | ICC |
|---|---|---|
| site | 15 | 0.006 |
## Confidence intervals for merMod models is an experimental feature. The
## intervals reflect only the variance of the fixed effects, not the random
## effects.
## subunit emmean SE df asymp.LCL asymp.UCL
## Judean Highlands 2.90 0.0614 Inf 2.78 3.02
## Carmel 3.04 0.0602 Inf 2.92 3.16
## Galilee 2.49 0.0642 Inf 2.36 2.61
##
## Results are given on the log (not the response) scale.
## Confidence level used: 0.95## [1] 0.7419288## [1] 0.154973## [1] 0.5081987| contrast | estimate | SE | df | z.ratio | p.value |
|---|---|---|---|---|---|
| Judean Highlands - Carmel | -0.1440769 | 0.0856729 | Inf | -1.681710 | 0.0926251 |
| Judean Highlands - Galilee | 0.4109161 | 0.0884443 | Inf | 4.646046 | 0.0000051 |
| Carmel - Galilee | 0.5549930 | 0.0878459 | Inf | 6.317801 | 0.0000000 |
significantly lower abundance in Galilee subunit compared to other two subunits (z-test post-hoc with fdr correction for 3 tests). No significant temporal trend.
Carmel and Judean highland plots have on average 8.929437 and 6.1163948 individuals more than Galilee plots, which is 74.1928811 and 6.1163948 percent higher, respectively.
community analysis using package MVabund
## Overlapping points were shifted along the y-axis to make them visible.##
## PIPING TO 2nd MVFACTOR## Only the variables Curruca.melanocephala, Streptopelia.decaocto, Turdus.merula, Parus.major, Garrulus.glandarius, Troglodytes.troglodytes, Chloris.chloris, Corvus.cornix, Alectoris.chukar, Carduelis.carduelis, Cinnyris.osea, Streptopelia.turtur were included in the plot
## (the variables with highest total abundance).
## Overlapping points were shifted along the y-axis to make them visible.##
## PIPING TO 2nd MVFACTOR## Only the variables Curruca.melanocephala, Streptopelia.decaocto, Turdus.merula, Parus.major, Garrulus.glandarius, Troglodytes.troglodytes, Chloris.chloris, Corvus.cornix, Alectoris.chukar, Carduelis.carduelis, Cinnyris.osea, Streptopelia.turtur were included in the plot
## (the variables with highest total abundance).
Mean-variance plot: there is a strong relationship, indicating that employing GLMs is the proper way to analyze, rather than OLS (assumption of homogeneity is violated). start model specification:
## nb po
## 544.1508 621.8585## [1] "POISSON"
## [1] "NEGATIVE BINOMIAL"
negative binomial model is better than poisson according to residuals and AIC comparison.
## nb po nb1
## 544.1508 621.8585 528.0333The addition of the explanatory variable ‘site’ is improving the AIC of the model. stepwise selection of model:
## Single term deletions
##
## Model:
## spp_no_rare ~ subunit * year_ct + cos_td_rad + sin_td_rad + site
## Df AIC
## <none> 7392.5
## cos_td_rad 14 7424.5
## sin_td_rad 14 7409.6
## site 196 7618.1
## subunit:year_ct 28 7428.0final model includes year, subunit, subunit X year, sampling time of year.
##
## Coefficients: (2 not defined because of singularities)
## wald value Pr(>wald)
## (Intercept) 4.673 0.033 *
## subunitCarmel 0.128 0.391
## subunitGalilee 8.317 0.211
## year_ct 9.078 0.039 *
## cos_td_rad 4.890 0.057 .
## sin_td_rad 4.764 0.081 .
## siteAmatzia 216.021 0.003 **
## siteBat Shlomo 0.433 0.255
## siteEitanim 3.908 0.177
## siteElyakim 0.361 0.256
## siteEshtaol 5.035 0.193
## siteGivat Yeshayahu 3.083 0.595
## siteKabri 183.169 0.010 **
## siteKerem Maharal 0.788 0.164
## siteManara 3.559 0.548
## siteMeron 13.655 0.116
## siteOfer 0.320 0.298
## siteRamat Hashofet Inf 0.339
## siteRamot Naftali 155.849 0.025 *
## siteZuriel 58.601 0.360
## subunitCarmel:year_ct 868.986 0.071 .
## subunitGalilee:year_ct 430.799 0.444
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Test statistic: NaN, p-value: 0.001
## Arguments:
## Test statistics calculated assuming response assumed to be uncorrelated
## P-value calculated using 999 resampling iterations via pit.trap resampling (to account for correlation in testing).## Analysis of Deviance Table
##
## Model: spp_no_rare ~ subunit * year_ct + cos_td_rad + sin_td_rad + site
##
## Multivariate test:
## Res.Df Df.diff Dev Pr(>Dev)
## (Intercept) 224
## subunit 222 2 225.4 0.01 **
## year_ct 221 1 27.3 0.02 *
## cos_td_rad 220 1 25.8 0.04 *
## sin_td_rad 219 1 26.7 0.04 *
## site 205 14 609.4 0.01 **
## subunit:year_ct 205 2 91.5 0.01 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Univariate Tests:
## Alectoris.chukar Carduelis.carduelis
## Dev Pr(>Dev) Dev Pr(>Dev)
## (Intercept)
## subunit 8.156 0.11 1.022 0.59
## year_ct 0.531 0.98 0.369 0.98
## cos_td_rad 0.33 0.98 0.889 0.97
## sin_td_rad 1.35 0.91 0.853 0.95
## site 22.31 0.18 32.744 0.03
## subunit:year_ct 1.025 1.00 0.821 1.00
## Chloris.chloris Cinnyris.osea Corvus.cornix
## Dev Pr(>Dev) Dev Pr(>Dev) Dev
## (Intercept)
## subunit 7.569 0.11 2.521 0.52 20.134
## year_ct 0.002 0.98 0.729 0.95 0.381
## cos_td_rad 1.148 0.90 0.491 0.98 0.388
## sin_td_rad 0.242 0.99 3.904 0.51 0.214
## site 32.49 0.03 43.967 0.01 68.936
## subunit:year_ct 26.268 0.01 0.329 1.00 0.346
## Curruca.melanocephala Garrulus.glandarius
## Pr(>Dev) Dev Pr(>Dev) Dev
## (Intercept)
## subunit 0.01 55.321 0.01 21.225
## year_ct 0.98 0.253 0.98 2.582
## cos_td_rad 0.98 0.141 0.98 0.015
## sin_td_rad 0.99 1.283 0.91 1.559
## site 0.01 112.609 0.01 36.36
## subunit:year_ct 1.00 6.213 0.44 6.028
## Parus.major Prinia.gracilis
## Pr(>Dev) Dev Pr(>Dev) Dev Pr(>Dev)
## (Intercept)
## subunit 0.01 7.029 0.13 13.377 0.02
## year_ct 0.75 0.79 0.95 2.431 0.75
## cos_td_rad 0.98 0.388 0.98 6.876 0.10
## sin_td_rad 0.91 0.071 0.99 0.063 0.99
## site 0.01 25.601 0.11 43.647 0.01
## subunit:year_ct 0.44 3.497 0.80 10.608 0.10
## Pycnonotus.xanthopygos Streptopelia.decaocto
## Dev Pr(>Dev) Dev Pr(>Dev)
## (Intercept)
## subunit 28.636 0.01 22.355 0.01
## year_ct 4.077 0.39 5.457 0.21
## cos_td_rad 6.935 0.10 1.756 0.81
## sin_td_rad 2.053 0.86 2.518 0.76
## site 28.335 0.04 44.621 0.01
## subunit:year_ct 4.756 0.55 2.621 0.91
## Streptopelia.turtur Troglodytes.troglodytes
## Dev Pr(>Dev) Dev Pr(>Dev)
## (Intercept)
## subunit 6.122 0.19 8.512 0.08
## year_ct 7.971 0.05 1.738 0.82
## cos_td_rad 1.918 0.81 0.423 0.98
## sin_td_rad 0.339 0.99 1.53 0.91
## site 37.084 0.01 66.751 0.01
## subunit:year_ct 19.455 0.01 1.953 0.92
## Turdus.merula
## Dev Pr(>Dev)
## (Intercept)
## subunit 23.453 0.01
## year_ct 0.014 0.98
## cos_td_rad 4.108 0.36
## sin_td_rad 10.699 0.03
## site 13.943 0.39
## subunit:year_ct 7.626 0.34
## Arguments:
## Test statistics calculated assuming uncorrelated response (for faster computation)
## P-value calculated using 99 iterations via PIT-trap resampling.Exmaine the effect of the interaction of year with subunit on individual species:
Chloris chloris - ירקון
## (Intercept) subunitCarmel subunitGalilee
## -0.96400244 0.50323313 -3.70807125
## year_ct cos_td_rad sin_td_rad
## 0.08755241 0.49374800 -1.40328595
## siteAmatzia siteBat Shlomo siteEitanim
## -0.43973951 0.22462235 -1.61670336
## siteElyakim siteEshtaol siteGivat Yeshayahu
## -0.03748978 -1.64159091 -0.49195031
## siteKabri siteKerem Maharal siteManara
## -11.48345220 0.21183637 2.12552585
## siteMeron siteOfer siteRamat Hashofet
## 0.90924651 -0.58380540 0.68806958
## siteRamot Naftali siteZuriel subunitCarmel:year_ct
## 2.80845388 1.93215468 -0.60521788
## subunitGalilee:year_ct
## 0.08947675## (Intercept) subunitCarmel subunitGalilee
## -0.96400397 1.19130135 -1.77591982
## year_ct cos_td_rad sin_td_rad
## 0.08755218 0.49375012 -1.40328949
## siteAmatzia siteBat Shlomo siteEitanim
## -0.43974002 -0.46344768 -1.61670474
## siteElyakim siteEshtaol siteGivat Yeshayahu
## -0.72555956 -1.64159205 -0.49195089
## siteKabri siteKerem Maharal siteManara
## -36.38941319 -0.47623381 0.19337109
## siteMeron siteOfer siteRamat Hashofet
## -1.02290839 -1.27187554 NA
## siteRamot Naftali siteZuriel subunitCarmel:year_ct
## 0.87629915 NA -0.60521784
## subunitGalilee:year_ct
## 0.08947719## Warning in predict.lm(object, newdata, se.fit, scale = residual.scale, type =
## if (type == : prediction from a rank-deficient fit may be misleading
## Warning in predict.lm(object, newdata, se.fit, scale = residual.scale, type =
## if (type == : prediction from a rank-deficient fit may be misleading
## Warning in predict.lm(object, newdata, se.fit, scale = residual.scale, type =
## if (type == : prediction from a rank-deficient fit may be misleading
## Warning in predict.lm(object, newdata, se.fit, scale = residual.scale, type =
## if (type == : prediction from a rank-deficient fit may be misleading
## Warning in predict.lm(object, newdata, se.fit, scale = residual.scale, type =
## if (type == : prediction from a rank-deficient fit may be misleading
## Warning in predict.lm(object, newdata, se.fit, scale = residual.scale, type =
## if (type == : prediction from a rank-deficient fit may be misleading## [1] 97.33152## [1] 0.9733152## [1] 245.2869## [1] 2.45287## [1] 84.57122## [1] 0.8457118## NOTE: A nesting structure was detected in the fitted model:
## site %in% subunit## subunit year_ct.trend SE df asymp.LCL asymp.UCL
## Judean Highlands 0.0876 0.0811 Inf -0.0714 0.246
## Carmel -0.5177 0.1432 Inf -0.7983 -0.237
## Galilee 0.1770 0.0817 Inf 0.0169 0.337
##
## Results are averaged over the levels of: site
## Confidence level used: 0.95## subunit year_ct.trend SE df z.ratio p.value
## Judean Highlands 0.0876 0.0811 Inf 1.080 0.2802
## Carmel -0.5177 0.1432 Inf -3.615 0.0009
## Galilee 0.1770 0.0817 Inf 2.167 0.0453
##
## Results are averaged over the levels of: site
## P value adjustment: fdr method for 3 testsStreptopelia turtur - תור מצוי
## Warning in glm.nb(formula = Streptopelia.turtur ~ subunit * year_ct +
## cos_td_rad + : alternation limit reached## (Intercept) subunitCarmel subunitGalilee
## -3.9898684 0.3221667 -8.6448468
## year_ct cos_td_rad sin_td_rad
## -0.2506289 3.6341962 -2.4036677
## siteAmatzia siteBat Shlomo siteEitanim
## 0.4862208 -9.6212214 -12.8578634
## siteElyakim siteEshtaol siteGivat Yeshayahu
## 1.4577399 -12.9830193 0.2301571
## siteKabri siteKerem Maharal siteManara
## -6.8003372 3.2842543 -6.9377785
## siteMeron siteOfer siteRamat Hashofet
## -6.7827383 2.9002185 2.3011754
## siteRamot Naftali siteZuriel subunitCarmel:year_ct
## 6.2037641 5.6722431 -1.0065992
## subunitGalilee:year_ct
## 0.4613248## (Intercept) subunitCarmel subunitGalilee
## -3.9900362 2.6233626 -2.9726488
## year_ct cos_td_rad sin_td_rad
## -0.2506312 3.6343556 -2.4038109
## siteAmatzia siteBat Shlomo siteEitanim
## 0.4861928 -34.3772736 -35.8295646
## siteElyakim siteEshtaol siteGivat Yeshayahu
## -0.8434370 -36.0054770 0.2301540
## siteKabri siteKerem Maharal siteManara
## -35.1184441 0.9830785 -35.2720810
## siteMeron siteOfer siteRamat Hashofet
## -35.1353042 0.5990394 NA
## siteRamot Naftali siteZuriel subunitCarmel:year_ct
## 0.5315188 NA -1.0066237
## subunitGalilee:year_ct
## 0.4613286## Warning in predict.lm(object, newdata, se.fit, scale = residual.scale, type =
## if (type == : prediction from a rank-deficient fit may be misleading## Warning in predict.lm(object, newdata, se.fit, scale = residual.scale, type =
## if (type == : prediction from a rank-deficient fit may be misleading
## Warning in predict.lm(object, newdata, se.fit, scale = residual.scale, type =
## if (type == : prediction from a rank-deficient fit may be misleading
## Warning in predict.lm(object, newdata, se.fit, scale = residual.scale, type =
## if (type == : prediction from a rank-deficient fit may be misleading
## Warning in predict.lm(object, newdata, se.fit, scale = residual.scale, type =
## if (type == : prediction from a rank-deficient fit may be misleading
## Warning in predict.lm(object, newdata, se.fit, scale = residual.scale, type =
## if (type == : prediction from a rank-deficient fit may be misleading## [1] 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 1 0 0 0 0 0 0 0 0
## [26] 0 0 0 0 0 1 0 21 0 0 0 0 0 0 0 0 0 0 0 0 0 2 2 0 0
## [51] 0 1 0 1 0 0 0 0 0 0 2 4 2 0 0 0 0 0 0 0 0 0 0 0 0## [1] 99.98494## [1] 0.9998494## [1] 245.2869## [1] 3.370521## [1] 478.0084## [1] 0.8269922## NOTE: A nesting structure was detected in the fitted model:
## site %in% subunit## subunit year_ct.trend SE df asymp.LCL asymp.UCL
## Judean Highlands -0.251 0.157 Inf -0.558 0.0572
## Carmel -1.257 0.395 Inf -2.032 -0.4826
## Galilee 0.211 0.164 Inf -0.111 0.5321
##
## Results are averaged over the levels of: site
## Confidence level used: 0.95## subunit year_ct.trend SE df z.ratio p.value
## Judean Highlands -0.251 0.157 Inf -1.596 0.1658
## Carmel -1.257 0.395 Inf -3.181 0.0044
## Galilee 0.211 0.164 Inf 1.285 0.1988
##
## Results are averaged over the levels of: site
## P value adjustment: fdr method for 3 testsThe effect of subunit interaction with year is not big. There are two species for which this interaction is significant: For Chloris chloris there are two subunits with opposing trends which are significantly different from zero. For Streptopelia turtur there is only one subunit with a temporal trend which is significantly different from zero.
Re-run model without subunit:year interaction and without site for simplification of the interpretation of results
## Single term deletions
##
## Model:
## spp_no_rare ~ subunit + year_ct + cos_td_rad + sin_td_rad
## Df AIC
## <none> 7645.4
## subunit 28 7793.5
## year_ct 14 7645.1
## cos_td_rad 14 7653.7
## sin_td_rad 14 7644.1drop sine
## Single term deletions
##
## Model:
## spp_no_rare ~ subunit + year_ct + cos_td_rad
## Df AIC
## <none> 7644.1
## subunit 28 7803.4
## year_ct 14 7645.5
## cos_td_rad 14 7641.9drop cosine
## Single term deletions
##
## Model:
## spp_no_rare ~ subunit + year_ct
## Df AIC
## <none> 7641.9
## subunit 28 7807.1
## year_ct 14 7641.2year should be dropped. keep year because difference in AIC is very small (<1) final model includes subunit and year.
##
## Test statistics:
## wald value Pr(>wald)
## (Intercept) 18.376 0.001 ***
## subunitCarmel 9.569 0.001 ***
## subunitGalilee 11.415 0.001 ***
## year_ct 5.502 0.019 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Test statistic: 16, p-value: 0.001
## Arguments:
## Test statistics calculated assuming response assumed to be uncorrelated
## P-value calculated using 999 resampling iterations via pit.trap resampling (to account for correlation in testing).## Analysis of Deviance Table
##
## Model: spp_no_rare ~ subunit + year_ct
##
## Multivariate test:
## Res.Df Df.diff Dev Pr(>Dev)
## (Intercept) 224
## subunit 222 2 225.43 0.01 **
## year_ct 221 1 27.32 0.02 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Univariate Tests:
## Alectoris.chukar Cinnyris.osea Prinia.gracilis
## Dev Pr(>Dev) Dev Pr(>Dev) Dev
## (Intercept)
## subunit 8.156 0.11 2.521 0.52 13.377
## year_ct 0.531 0.99 0.729 0.99 2.431
## Pycnonotus.xanthopygos Streptopelia.decaocto
## Pr(>Dev) Dev Pr(>Dev) Dev
## (Intercept)
## subunit 0.02 28.636 0.01 22.355
## year_ct 0.73 4.077 0.45 5.457
## Streptopelia.turtur Carduelis.carduelis
## Pr(>Dev) Dev Pr(>Dev) Dev Pr(>Dev)
## (Intercept)
## subunit 0.01 6.122 0.21 1.022 0.66
## year_ct 0.21 7.971 0.06 0.369 0.99
## Chloris.chloris Corvus.cornix
## Dev Pr(>Dev) Dev Pr(>Dev)
## (Intercept)
## subunit 7.569 0.15 20.134 0.01
## year_ct 0.002 0.99 0.381 0.99
## Curruca.melanocephala Garrulus.glandarius
## Dev Pr(>Dev) Dev Pr(>Dev)
## (Intercept)
## subunit 55.321 0.01 21.225 0.01
## year_ct 0.253 0.99 2.582 0.71
## Parus.major Troglodytes.troglodytes Turdus.merula
## Dev Pr(>Dev) Dev Pr(>Dev) Dev
## (Intercept)
## subunit 7.029 0.15 8.512 0.10 23.453
## year_ct 0.79 0.99 1.738 0.88 0.014
##
## Pr(>Dev)
## (Intercept)
## subunit 0.01
## year_ct 0.99
## Arguments:
## Test statistics calculated assuming uncorrelated response (for faster computation)
## P-value calculated using 99 iterations via PIT-trap resampling.Factors year and subunit have statistically significant effects on community composition.
## Loading required package: ggnewscale## Warning: Using `size` aesthetic for lines was deprecated in ggplot2 3.4.0.
## ℹ Please use `linewidth` instead.
## This warning is displayed once every 8 hours.
## Call `lifecycle::last_lifecycle_warnings()` to see where this warning was
## generated.## Warning: The `size` argument of `element_line()` is deprecated as of ggplot2 3.4.0.
## ℹ Please use the `linewidth` argument instead.
## This warning is displayed once every 8 hours.
## Call `lifecycle::last_lifecycle_warnings()` to see where this warning was
## generated.## [1] "black arrow is median of synanthrope/invasive\n\n"
## [1] "arrow is median of\nsynanthrope / invasive in black\n"
Results for individual species
Streptopelia turtur - תור מצוי
## Warning in glm.nb(formula = paste(spec_var_name, "subunit + year_ct", sep =
## "~"), : alternation limit reached## (Intercept) subunitCarmel subunitGalilee year_ct
## -0.06200274 0.29046887 -0.57347255 -0.28736804## (Intercept) subunitGalilee year_ct
## 0.3230341 -0.6299812 -0.3306849## [1] "RESULTS FOR Streptopelia turtur"## [1] "P-values"## term p
## 1: (Intercept) NA
## 2: subunit 0.21
## 3: year_ct 0.06
## [1] "change in Streptopelia turtur abundance over monitoring period: -90.1213519975731 %"Corvus cornix - עורב אפור
## (Intercept) subunitCarmel subunitGalilee year_ct
## -1.92332039 1.89209666 1.73679961 -0.03860959## (Intercept) subunitCarmel subunitGalilee
## -2.120264 1.897120 1.734601
## subunit emmean SE df asymp.LCL asymp.UCL
## Judean Highlands -2.120 0.385 Inf -2.875 -1.3652
## Carmel -0.223 0.232 Inf -0.678 0.2321
## Galilee -0.386 0.239 Inf -0.853 0.0818
##
## Results are given on the log (not the response) scale.
## Confidence level used: 0.95## [1] 17.70367## [1] 566.5867## [1] 466.3262## contrast estimate SE df z.ratio p.value
## Judean Highlands - Carmel -1.897 0.450 Inf -4.217 0.0001
## Judean Highlands - Galilee -1.735 0.453 Inf -3.828 0.0002
## Carmel - Galilee 0.163 0.333 Inf 0.488 0.6255
##
## Results are given on the log (not the response) scale.
## P value adjustment: fdr method for 3 testsCurruca melanocephala - סבכי שחור-ראש
## (Intercept) subunitCarmel subunitGalilee year_ct
## 1.59022736 -0.02680635 -1.02220624 -0.01100033## (Intercept) subunitCarmel subunitGalilee
## 1.53183667 -0.02332467 -1.02101104
## subunit emmean SE df asymp.LCL asymp.UCL
## Judean Highlands 1.532 0.0867 Inf 1.362 1.702
## Carmel 1.509 0.0871 Inf 1.338 1.679
## Galilee 0.511 0.1124 Inf 0.291 0.731
##
## Results are given on the log (not the response) scale.
## Confidence level used: 0.95## [1] 171.2851## [1] 2.326654## [1] 177.5969## contrast estimate SE df z.ratio p.value
## Judean Highlands - Carmel 0.0233 0.123 Inf 0.190 0.8495
## Judean Highlands - Galilee 1.0210 0.142 Inf 7.193 <.0001
## Carmel - Galilee 0.9977 0.142 Inf 7.016 <.0001
##
## Results are given on the log (not the response) scale.
## P value adjustment: fdr method for 3 testsGarrulus glandarius - עורבני
## (Intercept) subunitCarmel subunitGalilee year_ct
## 0.4338510 0.2170025 -0.9304102 -0.0648824## (Intercept) subunitCarmel subunitGalilee
## 0.1251631 0.1823216 -0.9461437
## subunit emmean SE df asymp.LCL asymp.UCL
## Judean Highlands 0.125 0.154 Inf -0.1769 0.427
## Carmel 0.307 0.148 Inf 0.0182 0.597
## Galilee -0.821 0.206 Inf -1.2240 -0.418
##
## Results are given on the log (not the response) scale.
## Confidence level used: 0.95## [1] 208.9471## [1] 19.96142## [1] 157.5387## contrast estimate SE df z.ratio p.value
## Judean Highlands - Carmel -0.182 0.213 Inf -0.854 0.3929
## Judean Highlands - Galilee 0.946 0.257 Inf 3.682 0.0003
## Carmel - Galilee 1.128 0.253 Inf 4.458 <.0001
##
## Results are given on the log (not the response) scale.
## P value adjustment: fdr method for 3 testsPrinia gracilis - פשוש
## (Intercept) subunitCarmel subunitGalilee year_ct
## -1.7682269 -2.6590856 -1.2099970 0.1569255## (Intercept) subunitCarmel subunitGalilee
## -0.8209806 -2.8033604 -1.1939225
## subunit emmean SE df asymp.LCL asymp.UCL
## Judean Highlands -0.821 0.366 Inf -1.54 -0.104
## Carmel -3.624 0.777 Inf -5.15 -2.102
## Galilee -2.015 0.451 Inf -2.90 -1.131
##
## Results are given on the log (not the response) scale.
## Confidence level used: 0.95## [1] 399.7811## [1] 1549.405## [1] 230.0256## contrast estimate SE df z.ratio p.value
## Judean Highlands - Carmel 2.80 0.858 Inf 3.266 0.0033
## Judean Highlands - Galilee 1.19 0.581 Inf 2.057 0.0596
## Carmel - Galilee -1.61 0.898 Inf -1.792 0.0731
##
## Results are given on the log (not the response) scale.
## P value adjustment: fdr method for 3 testsPycnonotus.xanthopygos - בולבול ממושקף
## (Intercept) subunitCarmel subunitGalilee year_ct
## -1.2783990 -2.7997326 -1.6689719 0.1565226## (Intercept) subunitCarmel subunitGalilee
## -0.3856625 -2.8332133 -1.6292405
## subunit emmean SE df asymp.LCL asymp.UCL
## Judean Highlands -0.386 0.243 Inf -0.863 0.0912
## Carmel -3.219 0.611 Inf -4.416 -2.0220
## Galilee -2.015 0.374 Inf -2.747 -1.2826
##
## Results are given on the log (not the response) scale.
## Confidence level used: 0.95## [1] 233.3424## [1] 1599.637## [1] 409.8773## contrast estimate SE df z.ratio p.value
## Judean Highlands - Carmel 2.83 0.657 Inf 4.310 <.0001
## Judean Highlands - Galilee 1.63 0.446 Inf 3.654 0.0004
## Carmel - Galilee -1.20 0.716 Inf -1.682 0.0926
##
## Results are given on the log (not the response) scale.
## P value adjustment: fdr method for 3 testsStreptopelia decaocto - תור צווארון
## (Intercept) subunitCarmel subunitGalilee year_ct
## 1.63780172 0.23468638 -0.55918725 -0.06401215## (Intercept) subunitCarmel subunitGalilee
## 1.3244190 0.2188792 -0.5667333
## subunit emmean SE df asymp.LCL asymp.UCL
## Judean Highlands 1.324 0.111 Inf 1.107 1.542
## Carmel 1.543 0.108 Inf 1.332 1.755
## Galilee 0.758 0.123 Inf 0.517 0.998
##
## Results are given on the log (not the response) scale.
## Confidence level used: 0.95## [1] 119.2407## [1] 24.48313## [1] 76.12081## contrast estimate SE df z.ratio p.value
## Judean Highlands - Carmel -0.219 0.155 Inf -1.414 0.1573
## Judean Highlands - Galilee 0.567 0.165 Inf 3.427 0.0009
## Carmel - Galilee 0.786 0.163 Inf 4.812 <.0001
##
## Results are given on the log (not the response) scale.
## P value adjustment: fdr method for 3 testsTurdus merula - שחרור
## (Intercept) subunitCarmel subunitGalilee year_ct
## 1.093806986 0.385184896 -0.271365662 0.002667568## (Intercept) subunitCarmel subunitGalilee
## 1.1074619 0.3861910 -0.2716584
## subunit emmean SE df asymp.LCL asymp.UCL
## Judean Highlands 1.107 0.0957 Inf 0.920 1.29
## Carmel 1.494 0.0880 Inf 1.321 1.67
## Galilee 0.836 0.1026 Inf 0.635 1.04
##
## Results are given on the log (not the response) scale.
## Confidence level used: 0.95## [1] 93.09266## [1] 47.25565## [1] 31.12751## contrast estimate SE df z.ratio p.value
## Judean Highlands - Carmel -0.386 0.130 Inf -2.971 0.0045
## Judean Highlands - Galilee 0.272 0.140 Inf 1.936 0.0528
## Carmel - Galilee 0.658 0.135 Inf 4.867 <.0001
##
## Results are given on the log (not the response) scale.
## P value adjustment: fdr method for 3 tests