Introduction

This document contains data analysis of Highly pathogenic avian influenza (HPAI) effects among colonial waterbirds in the lower Oder valley.

Highly pathogenic avian influenza (HPAI), effects among colonial waterbirds in the lower Oder valley.

Dominik Marchowski, Piotr Chara, Łukasz Borek, Zbigniew Kajzer, Szymon Bzoma

Data analysis

Loading Data

##   Code        site year        y        x dead total Mort_all Colony_Area
## 1    1    Kalensko 2023 52.63796 14.55681   58   266     10.9         250
## 2    2    Chlewice 2023 52.65511 14.49617   70    92     38.0         200
## 3    3    Bielinek 2023 52.95381 14.15164  203  1278      7.8        2330
## 4    5    Lubczyna 2023 53.50700 14.68948  150   544     13.8         500
## 5    6     Smiecka 2023 53.74997 14.50024    9   844      1.1       31500
## 6   10 DabieMarina 2023 53.40379 14.62535    0   223      0.0       95776
##   Habitat_Area DensBp.100m.2 hab1 hab2 DomSpec
## 1          250         106.0    A   AI      LS
## 2          200          46.0    A   AI      LS
## 3        28240          54.0    A   NI      LS
## 4       104700         108.0    N   NI      LS
## 5       260214           3.0    A   AI      LS
## 6       176975           0.2    N  NFV      CH

Data check for normality

## 
##  Shapiro-Wilk normality test
## 
## data:  hpai$DensBp.100m.2
## W = 0.76736, p-value = 0.003506

Data analysis

Spearman Correlation Analysis

## Warning: pakiet 'coin' został zbudowany w wersji R 4.4.1
## Ładowanie wymaganego pakietu: survival
## 
##  Approximative Spearman Correlation Test
## 
## data:  hpai$DensBp.100m.2 by hpai$Mort_all
## Z = 2.6629, p-value = 0.0024
## alternative hypothesis: true rho is not equal to 0
## 
##  Approximative Spearman Correlation Test
## 
## data:  hpai$Habitat_Area by hpai$DensBp.100m.2
## Z = -2.2423, p-value = 0.0187
## alternative hypothesis: true rho is not equal to 0
## 
##  Approximative Spearman Correlation Test
## 
## data:  hpai$Colony_Area by hpai$Habitat_Area
## Z = 2.8173, p-value = 5e-04
## alternative hypothesis: true rho is not equal to 0
## 
##  Approximative Spearman Correlation Test
## 
## data:  hpai$Colony_Area by hpai$DensBp.100m.2
## Z = -2.7311, p-value = 0.0012
## alternative hypothesis: true rho is not equal to 0
## 
##  Approximative Spearman Correlation Test
## 
## data:  hpai$Habitat_Area by hpai$Mort_all
## Z = -2.1774, p-value = 0.0242
## alternative hypothesis: true rho is not equal to 0

Generalized Linear Models (GLM)

## 
## Call:
## glm(formula = cbind(dead, (total * 2) - dead) ~ DensBp.100m.2, 
##     family = binomial, data = hpai)
## 
## Coefficients:
##               Estimate Std. Error z value Pr(>|z|)    
## (Intercept)   -3.68157    0.10202  -36.09   <2e-16 ***
## DensBp.100m.2  0.01894    0.00118   16.05   <2e-16 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 562.42  on 10  degrees of freedom
## Residual deviance: 269.63  on  9  degrees of freedom
## AIC: 310.5
## 
## Number of Fisher Scoring iterations: 5
## 
## Call:
## glm(formula = cbind(dead, (total * 2) - dead) ~ hab1, family = binomial, 
##     data = hpai)
## 
## Coefficients:
##             Estimate Std. Error z value Pr(>|z|)    
## (Intercept) -2.48387    0.05198 -47.786   <2e-16 ***
## hab1N        0.09809    0.09967   0.984    0.325    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 562.42  on 10  degrees of freedom
## Residual deviance: 561.47  on  9  degrees of freedom
## AIC: 602.34
## 
## Number of Fisher Scoring iterations: 5
## 
## Call:
## glm(formula = cbind(dead, (total * 2) - dead) ~ hab2, family = binomial, 
##     data = hpai)
## 
## Coefficients:
##             Estimate Std. Error z value Pr(>|z|)    
## (Intercept) -2.51721    0.07388 -34.072  < 2e-16 ***
## hab2NFV     -4.03815    1.00343  -4.024 5.71e-05 ***
## hab2NI       0.28473    0.09271   3.071  0.00213 ** 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 562.42  on 10  degrees of freedom
## Residual deviance: 441.05  on  8  degrees of freedom
## AIC: 483.92
## 
## Number of Fisher Scoring iterations: 6
## 
## Call:
## glm(formula = cbind(dead, (total * 2) - dead) ~ DomSpec, family = binomial, 
##     data = hpai)
## 
## Coefficients:
##             Estimate Std. Error z value Pr(>|z|)    
## (Intercept)   -6.555      1.001  -6.551 5.73e-11 ***
## DomSpecLS      4.211      1.002   4.204 2.62e-05 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 562.42  on 10  degrees of freedom
## Residual deviance: 450.67  on  9  degrees of freedom
## AIC: 491.54
## 
## Number of Fisher Scoring iterations: 6

Permutation Test Function

## predictor 
## 0.0189392
## [1] 0.1597633
## predictorN 
## 0.09808579
## [1] 0.9337349
## predictorNFV 
##    -4.038149
## [1] 0.006024096
## predictorLS 
##    4.211007
## [1] 0.006134969

Note that the echo = FALSE parameter was added to the code chunk to prevent printing of the R code that generated the plot.