Washington ? Justice Department prosecutions of international terrorism cases, which surged in volume after the Sept. 11, 2001 attacks, have nearly returned to the levels seen prior to the hijackings, according to an independent analysis of government data released Sunday.

The study of data by the Transactional Records Access Clearinghouse (TRAC), an affiliate of Syracuse University in New York, also showed that as many as nine out of 10 terrorism investigations do not result in prosecutions; that most charges are not related to terrorism; and that only about a third of those prosecuted end up in prison.

The findings, based on data compiled by the Justice Department’s Executive Office of U.S. Attorneys, echoes previous analyses by The Washington Post and others showing that most defendants in terrorism cases are charged with crimes unrelated to terrorism and many end up serving little or no prison time.

But the TRAC data also provide new insight into the extent to which prosecutions surged immediately after the Sept. 11 attacks, only to return to the far lower levels of previous years.

In 2002, federal prosecutors filed charges against 355 defendants in international terrorism cases, the study said. By last year, that number had dropped to 46, fewer than the number prosecuted in 2001. Just 19 such cases had been prosecuted so far this year, the study said.

The TRAC report called the decline in prosecutions “unexpected,” and said it seemed to contradict regular warnings from the government that the threat of global terrorism is higher than before.

“Considering the numerous warning statements from President Bush and other federal officials about the continuing nature of the terrorism threat … the gradual decline in these cases since the FY 2002 high point and the high rate at which prosecutors are declining to prosecute terrorism cases raises questions,” the report says.

But Justice Department spokesman Bryan Sierra said in a statement last week that the report contains “many flaws” and includes “misleading analysis.” He said sentencing statistics are not a good measure of the department’s counterterrorism efforts, and said it was not surprising that many referrals from the FBI or other investigative agencies would not result in prosecutions.

“This report does not take into account the significance of the Justice Department’s successful strategy of prevention through prosecution, which has helped protect this country from terrorists since the attacks of Sept. 11th,” Sierra said. “The department will continue to employ this strategy and work tirelessly to prevent another attack on U.S. soil.”

The dispute over the TRAC findings is the latest in a series of disagreements over the effectiveness of the Justice Department’s antiterrorism efforts, which can be measured in a number of different ways. The TRAC study, for example, is based on month-by-month data compiled by the U.S. attorneys’ office, and includes more than 6,400 referrals in “international terrorism” and “antiterrorism” cases since Sept. 11.

Atty. Gen. Alberto Gonzales and other senior Justice officials, however, have frequently relied on a different and much smaller list of terrorism-related cases compiled by the Criminal Division, which as of June included 441 defendants and 261 convictions.

An earlier Post analysis of that list found that most of the defendants were charged with crimes unrelated to terrorism and nearly half had no demonstrated connection to terrorist groups.

Many were explicitly cleared of terrorism links by the FBI or prosecutors, yet were still tallied as terrorism-related convictions by Justice.