Mixed model analysis

#1
I would greatly appreciate any help constructing a model to analyze the variation in strontium-to-carbon ratios in 204 fish otoliths (1 otolith per fish). I believe lmer is an appropriate R package, but am having difficulty implementing the model. The model includes terms to account for the nested and blocked design of field and laboratory sampling, and the possible influence of two covariates, fish body length and water temperature:

ratio = a + region + haul(region) + fish(haul(region)) + year + batch + length + temperature

where “a” is the intercept, and region is the main effect. Fish are nested in trawl haul (6 fish per haul), which is nested within region (6 hauls in each of 3 regions). For each fish (=otolith), 3 readings were obtained from different places on the otolith periphery. This design was repeated in two years, 2007 and 2011, which reflects data availability rather than some specifically chosen contrast (eg., warm vs. cold); thus, year is included as a block. Similarly, the otoliths were processed in batches, which is included as another block.

I recognize that this is a complicated problem, but any thoughts are greatly appreciated. If this post is inappropriate for this forum, please let me know and I'll take it down right away. I haven't posted on forums before. Thanks in advance for any suggestions, comments, or guidance.
 
Last edited: