Documentation of the changes from version 2 to version 3

Correction of errors and extensions

• Corrected a typo in maximum acceptable rainfall.(Aaron Boone)
• The range of energy changes allowed for the snow pack was clearly underestimated. Now the range is larger and should accommodate the bing changes observed in spring.(Aaron Boone)
• Some model can produce small negative soil wetnesses. The former value at -0.1 was too small and it has now been raised to -0.2.(Aaron Boone)
• A variable was added for the snow intercepted by the vegetation. The Rhône-AGG experiment seemed to indicate that models with this process presented a better simulation of the snow amounts.(Aaron Boone)
• Two inconsistencies between the level of priority for Qst and ACond in the output list and the quality check list. They were found by Dr. Agata.

Introduction of diagnostics related to CO₂ cycle

• first list of variable was added for the carbon cycle. This is the list suggested for PILPS-C1 by Nicolas Viovy.