[sam74]

I thought my original problem might be a bug so I emailed the files to the HEC RAS team. I actually got a reply the same evening. It took us a while to get on the same page but I think it is a bug. It appears after several manipulations and scenarios that HEC RAS has trouble with new steady flow input downstream of an optimized lateral structure. I originally thought it might have something to do with the fact that I had the three lateral structures so close but if I delete the steady flow inputs in the channels downstream of them and allow only the flow calculated by the optimization process to pass through I get pretty good looking numbers (constants). The channel continues on to a bridge crossing and on the other side of this (which is about 300 ft downstream of the last lateral structure) I had another steady flow input. Looking at the profile table on this section it did the same thing as it did in the midst of the lateral weirs. For instance my steady flow input for 1,2,5,10,25,50,100, and 500 yr events was 154, 164, 166, 171, 186, 200, 199, and 213. The flows listed at this same cross section in the profile table were 154, 160, 120, 80, 20, 0, 0, 0. I couldn't get the HEC RAS guy to admit to anything but he said they were working on this in their next release. He offered the advice that I break the area with multiple lateral weirs into three separate reaches and optimize them by hand. I did and this worked. Hope this helps someone