I would first like to thank EMI for the provision of the HMD dataset. 

When I was having a look at this data for one of my projects, I came up with a few questions:

1. For Lake Taupo inflow, there are 5 different datasets with different site numbers - Actual, Natural, Linnear, Non linear and Opearational. This is also the same for a few different flows. Am I right in assuming that all the different datasets represent the same flow using different methods? If that is the case, is there a most up-to-date method which EMI suggests using?

2. As I was trying to understand how the data was put together, I did some analysis on Lake Ohau (this flow was chosen because it had inflow, outflow, storage, and spill data). Below is a small sample of the analysis:

Lake Ohau analysis

The flow data is in reference to the left axis while the m3 quantity is along the right axis. The inflow is very closely correlated to the outflow. Also, the difference between the two has remained the same throughout the analyzed period at 12m3/s. Further, the spill is remaining the same at around 1.22 Mm3. Given these data, how is the storage data changing? When eyeballing it, it seems like when the inflow is increasing, there is an increase in storage (timeshifted by a few periods) but there is also an increase in outflow by the same amount.  Even the spill data is not exactly explaining the change. Can someone please clarify this?

3. I went through this thread for understanding how Specific_Energy is derived. There seem to be two different types of specific energy data - Scheme and Single stage. What is the difference between the two?

Thanks in advance.