Not long after National Guard trucks and an armada of small boats carried families out of swamped neighborhoods throughout the Baton Rouge area last August, local and federal officials pored over maps and pulled out calculators to produce jaw-dropping estimates of the scope of destruction.

A year later, with federal disaster claims verified and the long rebuilding process grinding on, a far-clearer picture has emerged of just how many people's homes saw murky water pour in.

Federal Emergency Management Agency data puts the number of flooded residences in the three hardest-hit parishes — East Baton Rouge, Livingston and Ascension — at 65,829. Local assessors in those parishes place the number of flooded structures, including homes, businesses and apartment buildings, at roughly 67,000.

Both figures suggest that initial estimates of the scope of the August 2016 flood significantly overstated the actual extent of damage — although it remains staggering — something disaster experts said is common as experts rely on chaotic reports and maps of high water in the immediate aftermath of disaster.

A definitive accounting of just how many people's homes, businesses or churches took on water — and the dollar-figure cost of the disaster and its aftermath — remains elusive. But a series of estimates and tabulations, compiled by different agencies for discreet purposes over the past year, have brought the margins much closer.

"The first numbers are really estimates, the second numbers are done by FEMA after inspections and are perhaps most accurate," said Craig Colten, a geographer and professor at Louisiana State University who studies disasters. "The third ones — inspections by local officials — tend to accommodate homeowners that want to stay in place and minimize their cost of doing that. None is entirely accurate and I don't know if there's ever going to be a full, accurate accounting."

Getting a handle on the flood's destruction — and how much remains uncovered by FEMA grants or flood insurance payouts — is essential when it comes to lobbying Congress for additional disaster recovery aid, according to several state officials.

And charting the spread of the flood can also be a key tool as parish and state officials push for flood-prevention projects and make planning decisions about potential new development.

Mark Harrell, Livingston Parish's director of Homeland Security and Emergency Preparedness, said he's been studying destruction in the Amite River watershed to build his case for major federal investment in projects to curb future floods. But Harrell and Livingston Parish President Layton Ricks said they've been frustrated with the lack of comprehensive data — and with delays and roadblocks in obtaining detailed information from FEMA.

"It's hard for us to make decisions on everything we do because we don't have the clear and precise data," Harrell said.

But those frustrations also highlight one of the key limitations of existing data on disaster damage — that government agencies and other groups invest in gathering data only when there's a compelling reason to collect it. Conducting an exhaustive census of flood damage would come with a hefty price tag, according to several experts, leaving gaps in the overall statistical picture.

"It's always a question of who's doing the counting and for what purpose," said Jack Heesch, a spokesman for FEMA.

Allison Plyer, executive director of the New Orleans Data Center, which has tracked changes in New Orleans since Hurricane Katrina, said there's simply no overarching purpose that would justify the huge cost of conducting a comprehensive study of flood damage.

"Curiosity questions are very expensive to answer. Do we need to know how many businesses flooded — who needs to know that? For what decision?" said Plyer, pointing to one of the fuzziest estimations of flood damage. "I think we assume all this information just exists but so much of it is based on individuals, households and businesses reporting the information to some government or private entity."

Residential damage

The clearest data on flood damage tracks the number of households whose homes took on water. That's in large part because of a range of government programs meant to help individuals — and homeowners in particular — find shelter and rebuild. Each program, state and federal, tracks numbers of applications and gathers data while doling out aid.

The resulting datasets don't provide an absolute, final total of affected homes — some eligible people invariably fail to apply, for instance — and determining the exact extent of damage across thousands of individual houses and apartments would be a huge undertaking, according to emergency response officials and experts who study disasters.

But numbers compiled by FEMA, for instance, do provide a fairly reliable measure of the flood's impact on families and individuals. The agency received a total of 153,407 applications for individual assistance for the August flood from Baton Rouge, Lafayette and the 20 other parishes included in the disaster declaration.

That figure includes people who lost property — including vehicles — but didn't see water come in their homes; multiple members of the same household who applied for aid; and residents who evacuated from inundated neighborhoods and registered with FEMA only to find out later that their homes were spared.

Working from there, FEMA inspectors verified that a total of 92,263 households — including those living in houses, trailers and individual apartments — suffered at least some flood damage. Renters account for a quarter of those households, or 23,383, while homeowners made up the rest.

In East Baton Rouge Parish, federal inspectors found 37,167 households suffered some degree of flood damage. The total tracks closely with an estimate generated by the City-Parish of 41,107 flooded residential structures — a number that also includes unoccupied or second homes — calculated through sophisticated mapping. The City-Parish's estimate includes the city of Central, but excludes Zachary and Baker, where an estimated 2,890 buildings of all types flooded.

In smaller Livingston Parish, FEMA found an eye-popping 20,808 households hit by the flood, or about half of the homes in the parish. That's down significantly from rough early estimates of the flood's damage, based primarily on tallying the number of buildings in flooded areas, but still represents an astounding number for a largely residential parish.

Over the past year, parish assessors have also settled on what they say are accurate tallies on the number of buildings damaged in the floods. Totals provided by the assessors in East Baton Rouge, Livingston and Ascension parishes roughly aligned with the FEMA-verified loss statistics.

Brian Wilson, the East Baton Rouge assessor, found about 41,000 flooded buildings in his parish. The City-Parish's mapping-based estimate tallied a total of 48,383 — or 30 percent of buildings in the parish. Both those figures consider multi-residence structures like apartment buildings as a single unit.

In Livingston Parish, Assessor Jeff Taylor found about 19,500 buildings flooded in August, a figure that accounts for roughly half of the parish's homes. That figure counts multi-unit apartment buildings as a single building. Splitting buildings out into individual flooded units, Taylor said, lifts the total to about 24,000 for the parish — though that number also adds in individual storage units.

In particularly hard-hit Denham Springs, Taylor said, roughly 72 percent of buildings flooded. That's significantly lower than some initial estimates of over 90 percent — but still a stunning figure for a single community.

Livingston Parish officials say they don't know how many people remain out of their homes — or simply walked away without rebuilding — but continue to hear regularly from folks deciding whether to rebuild or inquiring about debris pickup.

"It tells you some folks are just now starting to gut their homes," said Ricks, the parish president.

The differences between how assessors and FEMA count apartments — the federal agency counts each household, while assessors generally look to send the landlord a single tax bill — accounts for the gap between the numbers in Ascension Parish, said Assessor M.J. "Mert" Smiley Jr., whose office went building-to-building after the flood to determine the damage.

FEMA reported 7,854 households that suffered verified damage in Ascension Parish while Smiley said his office totaled 6,170 flood-damaged structures. Trailers parked behind a relative's home often aren't counted separately, either, Smiley said.

"If you have an apartment building with 20 units, I consider that one parcel. They consider that 20 parcels," Smiley said.

The latest survey of homes in Ascension Parish, completed a week ago, found 66 percent of people whose homes suffered damage in the flood have moved back in, Smiley said. That's good news for the families that have made it home — but it also points to a frustratingly slow pace of recovery for a substantial portion of the population.

"It's still worse than what people imagine," Smiley said. For many, the assessor said, financial struggles are continuing to hold up the recovery.

"Either the mortgage companies are holding the money or the insurance companies are making them ridiculous offers, they're having to go to court or hire lawyers to fight."

Business tallies elusive

Reliable information on the number of businesses is much harder to come by, largely because the primary type of government aid — loans from the federal Small Business Administration — are far less popular. A total of 4,735 flooded businesses applied to SBA across the disaster area, but state and parish-level officials say it's nearly impossible to say how many flooded businesses carried on without loans or simply folded.

A study of the August flood's economic impact commissioned by Louisiana Economic Development and released in September estimated that roughly 6,100 businesses flooded. Pat Forbes, director of the state Office of Community Development, said the study remains the best guess at the impact — but added that officials have "struggled mightily" at getting data on business damage.

"The people who go out of business don't apply for SBA, there are whole other groups of people who don't apply," Forbes said.

For churches, tracking down reliable data on the flood's damage is likewise challenging. Although most other types of nonprofits are eligible for federal disaster aid, religious institutions are generally excluded, with narrow exceptions for church-run facilities that serve non-religious functions, such as schools or community centers.

Without government aid on offer, there's no centralized application process that might provide an indication of how many churches went under water.

The state's Restore Louisiana program has been trying to reach out to flooded residents — even going door-to-door in once-flooded neighborhoods — in order to deliver more than $1 billion in federal recovery aid to homeowners of houses damaged in both the August disaster and the March flood that was particularly destructive in north Louisiana. Of the 57,923 households with the worst categories of damage from both floods, more than 63 percent were not covered by flood insurance. 

Central to the effort is a quick survey, available online at www.restore.la.gov, intended to determine what additional aid the government can deliver flood victims beyond FEMA grants, SBA loans and flood insurance payouts.

Forbes, who's heading up the program, said that without a greater response to the survey, it's difficult for his office to identify exactly who's immediately eligible for additional aid, how to direct additional money most effectively or if the program needs to be retooled. 

Among the key questions the Restore Louisiana survey hopes to assess is the actual cost of home repairs or rebuilding for homeowners hit by the flood and how much of that bill has gone uncovered.

Figuring out the dollar total of the gaps, Forbes said, would also allow Louisiana leaders to demonstrate to Congress how much more federal money is needed to effectively rebuild.

"We want to target programs and funding at the greatest unmet need. Maybe we've missed something we should've been paying attention to. We want to better understand people's losses in particular categories, whether they're accessing the program, if they have insurance, if they don't," he said. "All of that is information that helps us tailor programs and funding, too. It's not just about asking for more money. It's also knowing what to do with any more money."

Follow Bryn Stole on Twitter, @BrynStole.