Design of the Month

Prairie grasslands I

The following quotation is taken from ``Loss of plant species after chronic low-level nitrogen deposition to prairie grasslands'' by Christpher M. Clark and David Tilman, Nature 451 (2008) 712-715, doi:10.1038/nature06503 .

Our nitrogen-addition experiment was performed in two Minnesota prairie-like successional grasslands (abandoned from agriculture 1957 and 1934 respectively) and in a native savanna grassland. ... Plots received annual wet nitrogen deposition of approximately 6 kg ha-1 yr-1 (58% NH4, 42% NO3) and fertilizer nitrogen at 0, 10, 20, 34, 54 or 95 kg ha-1 yr-1 from 1982 to 2004. There were six replicates in each successional grassland and five replicates in the native savanna grassland (total 102 plots). ... An additional site adjacent to one successional grassland was initially disk ploughed and had 36 plots receiving the same six nitrogen addition treatments as above (n = 6) for a decade (1982-1991). ... From 1991 onwards, all treatments were stopped for a randomly selected half of the six replicates per treatment in this additional site to observe recovery (n =3). We measured plant species number ... in each plot every year from 1982 to 1994.

Ignoring the additional site:

  1. What were the experimental units, and how many were there?
  2. What were the observational units, and how many were there?
  3. What were the treatments, and how many were there?
  4. Draw the Hasse diagram for all relevant factors on the observational units, ignoring all treatment factors.
  5. Draw the Hasse diagram for all relevant factors on the set of treatments.
  6. Write down the skeleton analysis of variance, showing strata, sources and degrees of freedom.

Other designs

Page maintained by R. A. Bailey