Chapter 8 Raw data modelling (LMM) using elevation

#Create linear model formula
formula4 <- formula(o2_normalized ~ time_point*temperature+elevation+(1|individual)) #QC_normalized=o2 consumption normalized (ml/s/g)

#Fit the linear model
model4 <- lmer(formula4, data = calotriton_resp1_subset)
boundary (singular) fit: see help('isSingular')
#Print the model summary
summary(model4)
Linear mixed model fit by REML ['lmerMod']
Formula: o2_normalized ~ time_point * temperature + elevation + (1 | individual)
   Data: calotriton_resp1_subset

REML criterion at convergence: -681.1

Scaled residuals: 
     Min       1Q   Median       3Q      Max 
-1.94738 -0.63126  0.07569  0.73963  2.66971 

Random effects:
 Groups     Name        Variance  Std.Dev. 
 individual (Intercept) 0.000e+00 0.0000000
 Residual               1.194e-07 0.0003455
Number of obs: 58, groups:  individual, 31

Fixed effects:
                            Estimate Std. Error t value
(Intercept)                7.992e-04  1.060e-04   7.540
time_point2                2.789e-05  1.334e-04   0.209
temperature20              1.884e-04  1.266e-04   1.488
elevationLow               9.820e-05  9.107e-05   1.078
time_point2:temperature20 -5.198e-05  1.825e-04  -0.285

Correlation of Fixed Effects:
            (Intr) tm_pn2 tmpr20 elvtnL
time_point2 -0.640                     
temperatr20 -0.661  0.509              
elevationLw -0.491  0.075  0.051       
tm_pnt2:t20  0.476 -0.733 -0.696 -0.071
optimizer (nloptwrap) convergence code: 0 (OK)
boundary (singular) fit: see help('isSingular')
MuMIn::r.squaredGLMM(model4)
            R2m        R2c
[1,] 0.07199619 0.07199619

8.1 Raw data modelling (LMM) with interaction

#Create linear model formula
formula5 <- formula(o2_normalized ~ time_point*temperature*elevation+(1|individual)) #QC_normalized=o2 consumption normalized (ml/s/g)

#Fit the linear model
model5 <- lmer(formula5, data = calotriton_resp1_subset)
boundary (singular) fit: see help('isSingular')
#Print the model summary
summary(model5)
Linear mixed model fit by REML ['lmerMod']
Formula: o2_normalized ~ time_point * temperature * elevation + (1 | individual)
   Data: calotriton_resp1_subset

REML criterion at convergence: -641

Scaled residuals: 
     Min       1Q   Median       3Q      Max 
-2.13722 -0.68608  0.06061  0.56779  2.64762 

Random effects:
 Groups     Name        Variance  Std.Dev.
 individual (Intercept) 0.000e+00 0.00000 
 Residual               1.156e-07 0.00034 
Number of obs: 58, groups:  individual, 31

Fixed effects:
                                         Estimate Std. Error t value
(Intercept)                             0.0006666  0.0001388   4.802
time_point2                             0.0001416  0.0001892   0.748
temperature20                           0.0004667  0.0001836   2.541
elevationLow                            0.0003303  0.0001836   1.798
time_point2:temperature20              -0.0003454  0.0002584  -1.337
time_point2:elevationLow               -0.0001911  0.0002637  -0.725
temperature20:elevationLow             -0.0005235  0.0002503  -2.092
time_point2:temperature20:elevationLow  0.0005462  0.0003597   1.518

Correlation of Fixed Effects:
            (Intr) tm_pn2 tmpr20 elvtnL tm_2:20 tm_2:L tm20:L
time_point2 -0.734                                           
temperatr20 -0.756  0.555                                    
elevationLw -0.756  0.555  0.571                             
tm_pnt2:t20  0.537 -0.732 -0.711 -0.406                      
tm_pnt2:lvL  0.527 -0.718 -0.398 -0.697  0.525               
tmprtr20:lL  0.555 -0.407 -0.734 -0.734  0.522   0.511       
tm_pn2:20:L -0.386  0.526  0.511  0.511 -0.718  -0.733 -0.696
optimizer (nloptwrap) convergence code: 0 (OK)
boundary (singular) fit: see help('isSingular')
MuMIn::r.squaredGLMM(model5)
          R2m      R2c
[1,] 0.139987 0.139987