Northern Pintail Homework

Author

Katherine Cantrall

Read in data and necessary packages

duck<-read.csv("DuckData.csv", 
               stringsAsFactors = T)
require(tidyverse)
Loading required package: tidyverse
── Attaching packages ─────────────────────────────────────── tidyverse 1.3.2 ──
✔ ggplot2 3.4.0      ✔ purrr   0.3.4 
✔ tibble  3.1.8      ✔ dplyr   1.0.10
✔ tidyr   1.2.1      ✔ stringr 1.4.1 
✔ readr   2.1.2      ✔ forcats 0.5.2 
── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
✖ dplyr::filter() masks stats::filter()
✖ dplyr::lag()    masks stats::lag()

#Part 1

ggplot(duck, aes(x=State, y=NOPI))+
  geom_boxplot()+
   labs(y="Northern Pintail Abundance", title = "Northern Pintail Population Trends in Atlantic Flyway")
Warning: Removed 2 rows containing non-finite values (`stat_boxplot()`).

#Part 2

ggplot(duck, aes(x=Year, y=NOPI))+
  geom_point()+
  geom_smooth(method=lm)
`geom_smooth()` using formula = 'y ~ x'
Warning: Removed 2 rows containing non-finite values (`stat_smooth()`).
Warning: Removed 2 rows containing missing values (`geom_point()`).

#Part 3

#NOPI~Year
year<-lm(NOPI~Year, data=duck)
summary(year)

Call:
lm(formula = NOPI ~ Year, data = duck)

Residuals:
   Min     1Q Median     3Q    Max 
-15159 -11998  -9967   7431  94211 

Coefficients:
             Estimate Std. Error t value Pr(>|t|)
(Intercept) 140788.47  152325.51   0.924    0.357
Year           -63.90      76.73  -0.833    0.406

Residual standard error: 18430 on 180 degrees of freedom
  (2 observations deleted due to missingness)
Multiple R-squared:  0.003838,  Adjusted R-squared:  -0.001696 
F-statistic: 0.6936 on 1 and 180 DF,  p-value: 0.406
#-0.1%
#NOPI~State
state<- lm(NOPI~State, data=duck)
summary(state)

Call:
lm(formula = NOPI ~ State, data = duck)

Residuals:
   Min     1Q Median     3Q    Max 
-20624  -4690  -1956   1264  73447 

Coefficients:
            Estimate Std. Error t value Pr(>|t|)    
(Intercept)     5315       1613   3.295  0.00118 ** 
StateNC        27669       2272  12.180  < 2e-16 ***
StateVA        -1954       2272  -0.860  0.39071    
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

Residual standard error: 12490 on 179 degrees of freedom
  (2 observations deleted due to missingness)
Multiple R-squared:  0.5446,    Adjusted R-squared:  0.5395 
F-statistic:   107 on 2 and 179 DF,  p-value: < 2.2e-16
#53.9%
#NOPI~State+Year
yrsta<-lm(NOPI~Year+State, data=duck)
summary(yrsta)

Call:
lm(formula = NOPI ~ Year + State, data = duck)

Residuals:
   Min     1Q Median     3Q    Max 
-20447  -4776  -1529    921  75039 

Coefficients:
             Estimate Std. Error t value Pr(>|t|)    
(Intercept) 122426.75  103233.31   1.186    0.237    
Year           -58.98      51.99  -1.135    0.258    
StateNC      27636.62    2269.97  12.175   <2e-16 ***
StateVA      -1986.98    2269.97  -0.875    0.383    
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

Residual standard error: 12480 on 178 degrees of freedom
  (2 observations deleted due to missingness)
Multiple R-squared:  0.5479,    Adjusted R-squared:  0.5403 
F-statistic:  71.9 on 3 and 178 DF,  p-value: < 2.2e-16
#54%
#NOPI~State*Year
yrsti<-lm(NOPI~Year*State, data=duck)
summary(yrsti)

Call:
lm(formula = NOPI ~ Year * State, data = duck)

Residuals:
   Min     1Q Median     3Q    Max 
-21362  -5624   -820   2307  66810 

Coefficients:
               Estimate Std. Error t value Pr(>|t|)    
(Intercept)   6.185e+05  1.670e+05   3.703 0.000285 ***
Year         -3.088e+02  8.411e+01  -3.671 0.000320 ***
StateNC      -1.073e+06  2.391e+05  -4.490 1.28e-05 ***
StateVA      -4.133e+05  2.391e+05  -1.729 0.085572 .  
Year:StateNC  5.546e+02  1.204e+02   4.606 7.85e-06 ***
Year:StateVA  2.072e+02  1.204e+02   1.720 0.087124 .  
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

Residual standard error: 11850 on 176 degrees of freedom
  (2 observations deleted due to missingness)
Multiple R-squared:  0.5973,    Adjusted R-squared:  0.5858 
F-statistic:  52.2 on 5 and 176 DF,  p-value: < 2.2e-16
#58.5%

Data Interpretation

Part 1

This boxplot displays the Northern Pintail trends in the Atlantic flyway for 3 states; Maryland, North Carolina, and Virginia. Based on this plot, the greatest abundance of Northern Pintails can be found in North Carolina.

Part 2

This scatterplot displays the Northen Pintail population trend over the years in regards to the three previously mentioned states. Both Maryland and Virginia have had a downwards trend in population of the Northern Pintail while in North Carolina there has been a population increase. Maryland has a stronger downward trend in population than Virginia.

Part 3

Four models were run in determining which variables explain the most variance in the Northern Pintail population. Of these models were year, state, an additive model of the two, and an interactive model of the two. The worst preforming model was Year with an r-squared value of -0.001 meaning that -0.1% of the variance in Northern Pintail population can be explained by state. The state and additive model preformed very similarly. The best model was the interactive model. This model shows that 58.5% of the variation in Northern Pintail population can be explained by state*year.

Canvas Questions

  1. Which of the three models best represents the data? Justify your answer based on what we covered in class. The best model was the interactive model. This model shows that 58.5% of the variation in the Northern Pintail population can be explained by state*year. There is also a significant p-value of less than 0.05. This contrasts the worst-performing model that shows -0.1% of the variation in the NOPI population can be explained by year with a p-value of 0.406 which is not a significant value.

  2. Should the harvest limit for Northern Pintail be the same across the three states? Justify your answer. Even though the state model was not the best-performing mode, it did explain 53.9% of the variation in the Northern Pintail population. The best-performing interactive model tells us that the state is a defining parameter for population trends. Because of this, I do believe that the harvest limit for Northern Pintail should be different across the three states. This can also be supported by the plot in part two, where each state has evident different trends in the state’s population of Northern Pintails.

It looks like there was a steep population decline prior to 1970 that then levelled off. Let’s compare this observation from the Atlantic Flyway to we can learn from the Pintail entry from the Bird of the World Database through the VCU library (you have to use your VCU eID and PW). Specifically, read the following sections:

Demography and Populations -> Population Status + Population Regulation Conservation and Management -> Management

  1. Does this information corroborate what we see in our dataset for the Atlantic Flyway? Provide evidence to support your answer. There were record low numbers during extensive prairie drought in 1988–1991. There was a decline in the number of breeding birds. From 1975-1992 there was an 11% decline in Canadian prairies, a 7% decline in the U.S. prairies, and a 3% decline in Canada, but it remained stable in Alaska. This can be reflected in the Atlantic Flyway data set for the states of Virginia and Maryland, where there was a decrease in population. One explanation for the downward trends we see in the Atlantic Flyway is that during severe prairie droughts and low continental populations, 42-58% of the NOPI populations were found in Alaska.

  2. What are the primary limiting factors (things regulating NOPI populations)? High populations rely on abundant shallow wetlands. On top of this, many human activities affect the Northern Pintail population. Among these are shooting and trapping, pesticides and contaminants, plastics and lead, collisions, degradation of habitat, and disturbance of nest/roosting sites.

  3. What actions are being taken to bolster NOPI populations? Restoring wetlands and integrating waterfowl management with farming operations in the Prairie Pothole Region of Canada and the U.S. are two significant actions to bolster Northern Pintail populations. Several public and private organizations are working to protect and enhance breeding and wintering habitats through cooperative programs with public and private landowners. Another thing that will benefit NOPI populations is Conservation-oriented agriculture policy programs.