Which isn’t to say that reporters should ignore signs of recovery or trump up negative impacts where they don’t exist, but the media must avoid falling into the “mission-accomplished” trap until government, industry, and independent scientists have collected more data. Avoiding alarmism on one hand and complacency on the other is a difficult job, however. Take a controversial report that Michael Grunwald penned for Time magazine’s Web site on Thursday, arguing that “while it’s important to acknowledge that the long-term potential danger [of the spill] is simply unknowable for an underwater event that took place just three months ago – it does not seem to be inflicting severe environmental damage.”

Grunwald argues that the media have an “obvious incentive to accentuate the negative in the Gulf … because disasters drive ratings and sell magazines.” That may be true to a certain extent, but, as Mother Jones’s Kate Sheppard fairly points out, “if he’s going to criticize folks for making premature doomsday predictions, then he, too, shouldn’t engage in making preemptive declarations that the problem is exaggerated, either.”

Indeed, Grunwald’s piece sparked a bit of disagreement at Time, where his colleague, Bryan Walsh, felt compelled to offer this response:

I think it’s far too early to declare the oil spill a bust… The truth is we know very little about what the release of tens of millions of gallons of oil underwater will do to the marine ecosystems of the Gulf. Add in the application of some 2 million gallons of chemical dispersants, which have never been used—and were never meant to be used—in such vast quantities…

Grunwald is taking a land-centric view of the spill—as long as the oil doesn’t show up on the beach, it’s probably not doing much damage. But that’s far from clear today, just a little over 100 days from the sinking of the Deepwater Horizon.

To be fair, Grunwald doesn’t seem be “declaring victory” to the extent that Sheppard and Walsh imply, but in his effort to discourage alarmism definitely risks encouraging complacency. As The New York Times pointed out in an excellent survey of the long-term impacts oil spills in four other regions of globe, “Only 20 years ago, the conventional wisdom was that oil spills did almost all their damage in the first weeks, as fresh oil loaded with toxic substances hit wildlife and marsh grasses, washed onto beaches and killed fish and turtles in the deep sea.” But “hidden damage can last for years.”

The Times was careful to stress that “scientists … say the picture in the gulf is far from hopeless.” More journalists should emphasize that important fact, and Grunwald tried. But rather than complain about “overblown eco-fears,” he should have taken a cue from two recent articles from Reuters and The Washington Post, which highlighted the resiliency of marsh grasses and their importance to wetlands recovery without suggesting that we have nothing to worry about.

Indeed, until we can account for where all of the spill oil has gone, what was true at the beginning of the spill has never been more important: what is out of sight should never be out of mind. The crude on the surface may be “missing,” but to leave it at that misses the point.

Curtis Brainard is the editor of The Observatory, CJR's online critique of science and environment reporting. Follow him on Twitter @cbrainard.