States, Counties Find Fault with Bay Diet, Too

Up until recently, it's been the agricultural interests who have objected most to the requeirments of the new "Bay Diet", and faulted the model used for not having the right data.  Now states and counties around the Bay are examining what changes the "diet" means in their lifestyle, and some of them are not so happy either:

Virginia and several other states - including Maryland, Pennsylvania and Delaware - are complaining that a newly tweaked version of the model, known as 5.3.2, is leading to some weird and incomprehensible results for what local governments are expected to accomplish in the coming years to dramatically improve water quality by 2025.

In James City County, for example, data stemming from the previous model urged the county near Williamsburg to reduce nitrogen from farms, streets, storm drains and development sites by 8 percent, phosphorus by 11 percent and sediment by 20 percent. The guidance worried local officials, unsure how they would pay for environmental improvements and controls to hit those targets.

However, computer runs performed by the state using the new model prescribe something completely different: no reductions needed for nitrogen, and a 20 percent surplus of phosphorus and a 350 percent cushion for sediment.

So you went from needing a big cleanup, to being able to sell pollution credits to a neighboring jurisdiction.  What's wrong with that?  If it's true, maybe nothing.  But if it's caused by bad modeling, it suggests that someone else is being stuck with the bill, and they are, quite rightly, reluctant to pony up.

While the James City County discrepancies are extreme - new data show that most Virginia localities have to do more, not less, to help save the Bay - state and local officials face a quandary: How exactly to proceed in the face of changing targets?

"What do we say to our localities? 'Well, we think that these practices we are asking you to implement might help you reach your goal, but we really don't know what that goal is and we aren't sure the money you spend to implement these practices will make any difference?' " said Doug Domenech, Virginia's secretary of natural resources, Gov. Bob McDonnell's top environmental official.

The EPA, environmentalists and some scientists concede that the modeling is imperfect and will continue to be updated and improved. But they also say the states are not required to be so precise in their calculations, and that no one asked them to break down data county by county, pound by pound of pollutants, for what they need to do to help the effort.

What's Fritz's Second Law?  All models are wrong, the only question is how much and in what direction.  Now the EPA want to gloss over the obvious errors at the smaller scale and make the states decide who pays the bill.

But if the model can't really give good numbers on the county scale, who's to say that it's any better on the watershed scale.  Smearing bad number over larger areas doesn't suddenly make them good.