A couple of people approached to ask whether the creek could simply be dug back out. I told them restoration would have to be designed carefully, especially with fill material compacted in the old channel and water now reaching foundations it hadn't reached before. I didn't promise an easy fix, because false reassurance would have made me no better than the person we were challenging. I told them one thing I knew was true: every dollar spent on a new system was a dollar they shouldn't commit until someone had established exactly why the original one failed.
That night Brenda posted a photograph of herself standing in the rain in polished boots under a clear umbrella. The caption called her grateful for the opportunity to stand in the mud with her people. In the background, just outside the crop, was a street where homeowners were still hauling ruined drywall to the curb. Clare looked at the picture on my phone, shook her head, and asked me to stop showing her Brenda's posts for the evening.
I set the phone down. Outside, the creek was still silent, and far below the house the first mosquitoes of the warm season were gathering above pools that had formed where clean moving water once carried them away. Brenda had called the flowing creek a breeding ground. She had managed to turn it into a stagnant one.
My grandfather kept a black ledger because he believed memory was too generous to people who wanted to be right. I kept folders for much the same reason. By the time September turned toward October, the kitchen table held three kinds of documents: what the land had been, what Brenda had caused to happen to it, and the money that seemed to be moving behind the scenes. The categories were simple enough to explain to a neighbor and separate enough that each one could be checked by somebody who didn't know us.
I began with the maps. The United States Geological Survey topographic sheet showed Willow Branch descending from the spring in our hollow toward Cane Creek. County surveys and historical aerial photographs showed the same alignment, changing in small ways where the channel had wandered naturally but always remaining in the same valley bottom. The National Wetlands Inventory offered additional clues about the low ground, though it was not itself a final legal determination of federal jurisdiction.
That distinction mattered to me. I'd spent a career telling people the presence of blue ink on a map wasn't always the same thing as a formal legal finding, and I wasn't going to change my standards because the person accused of violating them had destroyed something I loved. Federal protection of streams can depend on current law and site-specific findings. So I pulled the actual records for Willow Creek's original development review and looked for a determination that addressed this particular branch rather than a general statement about local waterways.