Here's the Wilbur opinion. We're likely to see some significant differences here because Wilbur is more likely to give river channels that flow "too closely" to each other than would a physically-based simulation.
The first image is simply rivers. The second is the land.sea mask I used to pull the original coastline (there are some JPEG compression artifacts visible along the edges). The third image is the height field (I'm hoping the JPEG compression doesn't do too many awful things to it). The final one is the rendered color version of the data for reference.
The biggest issue with the incised terrain is the depths of the canyons and lack of flat terrain. If this were an island about the size of Hawaii then the incision might be plausible, but without a scale it's hard to know what sort of processing to apply. The physical processes that need to be simulated depend heavily on the size of each pixel (for example, most sediment/flow models are suitable for roughly meter-scale pixels with relatively uniform substrate characteristics, but most continental maps tend toward 100 meter or more per pixel and substrate characteristics are highly variable at that scale).
So, umm... What is the scale for this map or have I failed at reading once again?
I am ALWAYS amazed when I see the wonderful work both Redrobes and Waldonrate can do with their respective software apps!
I gave it a bit of a thrashing with the hose to try to get enough water to break the pool in the middle but it would not. I could have set the permeability up and it might have dried up eventually. If this were mine I would have added a drop to the terrain somewhere to force an outflow for it. Also I think generally theres too much high ground and my proggy was having a hard time with the sharp fall on the bottom left. But there we are you have the devs of Wilbur and GTS supplying maps of your land. Cant be bad huh ? Interesting for me as we haven't had a side by side comparison like this before.
EDIT -- Added a side view since its right here on my screen. This shows the problem with the large lake. Perhaps a giant push and it would overflow down towards the camera.
EDIT2 -- Looking back and comparing now to Wilburs I can see that it has done a better job than mine on the mountain top erosion. I like the pattern Wilbur is generating. Its also found a path to the right and it clearly had a big dip there so if it managed that automatically then that's impressive too. I don't make any attempt to automatically fix up basins, the program relies on it overflowing at some point and driving a new channel out of it. Usually that results in it dumping a lakes worth of water over the lowlands and eroding large chunks of that away in the process. That sort of overflow did occur on this map in the top right local below sea level areas which then breached into the sea fixing that up.
Dangit.. not at all where I imagined the rivers would go. :/
I really like the second...it's stunning. label it like with cocktail sticks with flags on (I'm serious) and it would really be fantastic.