91L's structure gradually improved heading into the evening Sunday, and this trend has continued at a somewhat faster pace so far overnight. The low now no longer appears disorganized, but rather looks much more like a wanna-be TD again.

There are two main considerations as to 91L's future forecast track worth mentioning.

First, the exact 'center' of 91L is still a bit of a mystery. This is often the case with developing tropical systems, especially when they are not yet even in the TD state, as 91L has been. The best, albeit elongated, surface circulation remains a good bit onshore in the southwesternmost portion of the Bay of Campeche. However, the circulation just above the surface appears to have reconfigured itself further offshore during the past 24 hours, and this is also the location now officially being tracked:

We likely won't have a really good handle on where the exact center is until recon arrives on the scene later today, Monday. This initial position of the low could have outsized impacts on the accuracy of model forecast tracks and intensities down the road, so we'll want to start off with getting that right.

For now, assuming that the estimated center location is 'close enough' to correct, recent model runs shown below as plotted by SFWMD for 02/12, 02/18z and 03/00z strongly suggest a consensus that the low could stay mostly if not entirely offshore until a final or eventual landfall around Louisiana mid to late week. However, the track range, or de facto "Cone of Uncertainty," does cover hundreds of miles.

I've outlined in heavy black pen the range, with the individual members (e.g. GFS, HWRF, UKMET) within. Note: XTRP is extrapolating of the past several hours of actual track - but in this case it is not really valid as such, because rather than a traditional track towards the northeast, we have merely witnessed a center jump. So unless that trend were to continue, a movement more towards the northwest or northish should resume in the near term.