The FACE experiment is relatively expensive, making it difficult to include multiple replicates for each treatment. This presents a challenge: Are the final carbon cycling results due to the treatments themselves or the inherent variations among the sample points?

Does this require comparing with baseline data? How to do it?