Sturgeon catch per unit effort per river mile

Import data

setwd("C:\\Users\\Jack\\Documents\\AB\\PVA MS Sturgeon 2019")
dat <- read.csv("sturgeon_length_data.csv")
head(dat)

How many rows are there?

nrow(dat)

[1] 6683

Weight imported as a factor, but this should be numeric.

dat$weight <- as.numeric(as.character(dat$weight))

NAs introduced by coercion

In the Excel sheet, NA’s were entered as the character “.”, but this gets converted to NA correctly.

How many NA’s are there for weight?

sum(is.na(dat$weight))

[1] 4628

Check this sum against the original file sum of 4459.28198.

sum(dat$weight, na.rm = TRUE)

[1] 4459.282

The sum matches.

The data are from an Excel spreadsheet. The Excel date has an origin of December 30, 1899, and needs to be converted for R.

dat$date <- as.Date(dat$date, origin = "1970-01-01")

Now check the data set again.

head(dat)

Aggregate data

dat2 <- aggregate(dat$numberCaught, by = list(species = dat$species, riverMile = dat$riverMile), FUN = sum)
head(dat2)

plot(dat2$riverMile[dat2$species == "PALLID"], dat2$x[dat2$species == "PALLID"])

fitPallid <- lm(dat2$x[dat2$species == "PALLID"] ~ dat2$riverMile[dat2$species == "PALLID"])
summary(fitPallid)

Call:
lm(formula = dat2$x[dat2$species == "PALLID"] ~ dat2$riverMile[dat2$species == 
    "PALLID"])

Residuals:
    Min      1Q  Median      3Q     Max 
-2.3616 -1.6425 -0.5549  0.4432 13.6356 

Coefficients:
                                          Estimate Std. Error t value Pr(>|t|)    
(Intercept)                              2.4116642  0.5929325   4.067  0.00011 ***
dat2$riverMile[dat2$species == "PALLID"] 0.0008329  0.0008900   0.936  0.35214    
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Residual standard error: 2.499 on 81 degrees of freedom
Multiple R-squared:  0.0107,    Adjusted R-squared:  -0.001517 
F-statistic: 0.8758 on 1 and 81 DF,  p-value: 0.3521

No significant trend for pallid sturgeon.

plot(dat2$riverMile[dat2$species == "SHOVELNOSE"], dat2$x[dat2$species == "SHOVELNOSE"])

fitShovel <- lm(dat2$x[dat2$species == "SHOVELNOSE"] ~ dat2$riverMile[dat2$species == "SHOVELNOSE"])
summary(fitShovel)

Call:
lm(formula = dat2$x[dat2$species == "SHOVELNOSE"] ~ dat2$riverMile[dat2$species == 
    "SHOVELNOSE"])

Residuals:
    Min      1Q  Median      3Q     Max 
 -72.36  -31.21  -12.70    5.48 1079.78 

Coefficients:
                                              Estimate Std. Error t value Pr(>|t|)   
(Intercept)                                  -14.02203   18.78971  -0.746  0.45652   
dat2$riverMile[dat2$species == "SHOVELNOSE"]   0.07892    0.02663   2.964  0.00347 **
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Residual standard error: 101.5 on 173 degrees of freedom
Multiple R-squared:  0.04832,   Adjusted R-squared:  0.04282 
F-statistic: 8.783 on 1 and 173 DF,  p-value: 0.003468

There is a significant trend for shovelnose sturgeon, but this might only be due to the two extremely large observations far upstream.