> 
> salmon.mod1 <- lm(logtrap ~ water)
> summary(salmon.mod1)

Call:
lm(formula = logtrap ~ water)

Residuals:
    Min      1Q  Median      3Q     Max 
-3.7177 -0.8663  0.3814  1.3382  2.9111 

Coefficients:
            Estimate Std. Error t value Pr(>|t|)    
(Intercept) 0.655146   0.460642   1.422    0.162    
water       0.015643   0.001866   8.385 1.36e-10 ***
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 

Residual standard error: 1.675 on 43 degrees of freedom
Multiple R-squared: 0.6205,     Adjusted R-squared: 0.6117 
F-statistic: 70.31 on 1 and 43 DF,  p-value: 1.359e-10 

> plot(residuals(salmon.mod1),type="l")
> 
> library(car)
> durbinWatsonTest(salmon.mod1)
 lag Autocorrelation D-W Statistic p-value
   1       0.3255393      1.287346   0.016
 Alternative hypothesis: rho != 0
> 
> library(nlme)
> 
> salmon.mod2 <- gls(logtrap ~ water, correlation=corARMA(p=1), method='ML')
> summary(salmon.mod2) 
Generalized least squares fit by maximum likelihood
  Model: logtrap ~ water 
  Data: NULL 
       AIC      BIC    logLik
  173.3490 180.5757 -82.67451

Correlation Structure: AR(1)
 Formula: ~1 
 Parameter estimate(s):
      Phi 
0.4217852 

Coefficients:
                Value Std.Error  t-value p-value
(Intercept) 1.0903352 0.5405791 2.016976    0.05
water       0.0132779 0.0017986 7.382224    0.00

 Correlation: 
      (Intr)
water -0.686

Standardized residuals:
       Min         Q1        Med         Q3        Max 
-2.2068095 -0.6521012  0.1809419  0.8540458  1.8526576 

Residual standard error: 1.67204 
Degrees of freedom: 45 total; 43 residual