When the Army Corps of Engineers began evaluating what went wrong with New Orleans' levees during Hurricane Katrina, University of Texas at Dallas President David E. Daniel got his chance to peer into one of history's worst engineering failures.

Dr. Daniel, an engineer, chaired a 14-member panel formed by the American Society of Civil Engineers that examined the corps' review, making sure the official version was credible.

"They wisely decided it would be a really good idea to have an external group look over their shoulder to review their work," he said. "Perhaps they were concerned that no one would trust them otherwise."

The outside reviewers found that the Army engineers' postmortem was well-done. But what it revealed about decades of nearsighted planning, shoddy engineering and official complacency surprised Dr. Daniel.

"The one finding that really shocked me was the problems are everywhere," he said. "Serious, serious problems at almost every level – I had not expected that."

Dr. Daniel will present a special report on his group's findings at 6 tonight in the McDermott Suite on the fourth level of UT Dallas' McDermott Library.

The panel found deep management problems with the entire hurricane protection system for the New Orleans area. Despite the enormous stakes, nobody was in charge.

"The problem, essentially, is that there are about 10 different agencies with different levels of responsibility, but no one with any authority or responsibility for the entire system," Dr. Daniel said. "We called it a sometimes poorly connected set of pieces rather than a thoughtfully integrated system."

That's a lesson that applies to any organization with large and complex responsibilities, Dr. Daniel said, from how NASA decides whether to launch astronauts into space to how a nuclear power plant keeps the public safe.

"Time has clearly taught us that the management can be just as important as the engineering, in particular how critical decisions are made at critical times," Dr. Daniel said. "You imagine building a nuclear power plant with nobody in charge. A fundamental problem in New Orleans is that no one is in charge."

The state-run evacuation of Houston for Hurricane Rita put the need for accountability in stark relief, Dr. Daniel said. "Wouldn't it have been nice in New Orleans if the governor of the state of Louisiana had said, 'Well, the state's going to become responsible for and take charge of this.' " Instead, "the [Louisiana] governor did not have a strong role or strong voice."

Centralized authority over public safety planning probably won't come easily. "I don't think it's practical to think that all these dozen or so entities that are involved are suddenly going to cede all of their authority to some 'king of levees,' " Dr. Daniel said.

"I think at some point, you've got to lock these people in a room and say, 'Come up with a rational and reasonable plan to work together more effectively,' " and to speak about safety in a strong voice. "Fixing the engineering alone, without fixing the fundamental management dysfunction, will not necessarily solve the problem."

The Corps of Engineers' approach to its hurricane protection mission illustrated how gravely flawed the whole process was, Dr. Daniel said.

"There were about 50 sections of levee that failed, and 46 of those 50 sections failed because of water rushing over the top and eroding the levees," he said. But the Corps of Engineers said it could not design the levees to resist erosion from overtopping because that would imply that it was exceeding its congressional authorization – in other words, the corps claimed that "Congress never authorized storms like Katrina."

The lessons of Katrina extend to other places, too, in the need to shake off complacency and plan for bad events that will happen eventually, Dr. Daniel said. In North Texas, he said, that bad event could be a truly catastrophic drought.

"Years go by and we get a little bit drier period occasionally. And we forget that it is inevitable that an awful drought will hit this area. It just will happen. It's just a question of when it will happen."

Dr. Daniel said he also brought back a lesson for running a complex system such as a university.

"Leadership does count and management does matter," he said. "Planning by wishful thinking is not a very effective way to plan."

E-mail rloftis@dallasnews.com