I'm encountering an increasing number of modelling specifications that dictate a maximum pipe length of 1000m in H2OMap. Is there a technical reason for this or is it just a nominal value that has been carried through and adopted over time for (in)convenience? The models are reticulation networks and are not used for transient analysis (where min/max pipe lengths do come into play).

Smaller pipes would increase the number of nodes thus increasing simulation runtime. If i recall correctly the algorithmic complexity of H2OMap is O~(n^2 m) where n is input data and m is the number of edges. Being in sparsely populated Australia, some of the models I run are quite large geographically so it is not unusual to encounter GIS data where pipe lengths are significantly larger than 1km.