Wiesenberg et al. made several observations on N-alkanes but differences were not explained but indeed exist http://www.geo.unizh.ch/~mschmidt/downloads/wiesenberg_OG_2004.pdf, Bush and mcInerney point out that CPI variability  is also linked to climate but sphagmum mosses are different, https://isen.northwestern.edu/doc/pdf/news/scholarlypapers/McInerney_9oct13_geochimica.pdf . In all case the reason is a wax composition  change and research is still in progress as far I know, Bernard and Joubes presented the state of art in http://www.ncbi.nlm.nih.gov/pubmed/23103356, but in Arabidopsis. Probably it is an issue for H2/D2 discrimination.

The odd-over-even predominance or, in case of n-fatty acids and n-alcohols even-over-odd predominance, results from the way plants synthesise long-chain alkyl compounds. Starting with short-chain compounds enzymes (prolongase) attach acetate (acetic acid) to the molecule, which is a 2-carbon molecule. The ultimate chain length of an alkyl compound depends on the tissue or cell type it will be incorporated in, i.e. whether it is part of cuticular waxes, cutin or suberin, for example. I recommend reading some papers on alkyl compound biosynthesis. Samuels et al. (2008) is a good starting point, for example.

Samuels et al. (2008), Annual Review of Plant Biology. 59, 683–707

Correction re. my previous answer:

The enzyme adding 2 carbons to a short-chain alkyl compound is called elongase.

There's plenty of information on this on the web, as well, e.g.: http://www.llamp.net/?q=Elongase%20pathway

Even though you are using biomarkers to reconstruct the paleo-environment, I'm afraid, there's no way around delving deeper into biology if you want to understand and interpret your data properly.

http://www.llamp.net/?q=Elongase%20pathway

Guido Bongi & Jens Holtvoeth @ Thanks for your valuable comments.