The survey flight Zephyr conducted over the upper reaches of the obstruction gave him precise information about the dam's structure. He flew the eastern flow path three times at low altitude, reading the turbulence patterns in the water surface below him with the pressure sensors in his wings.
The flow was concentrated into a narrow high-velocity band along the eastern channel wall, running fast over a bed of exposed gravel that the accelerated current had already scoured clean of its normal silt covering.
The critical discovery came on his second pass, when he read a significant pressure differential in the air above the western edge of the obstruction. The clay mass was not uniform — there was a section near its center where the material was less consolidated, where the bank roots had formed a looser matrix that had incorporated more air pockets during the collapse.
This section had a different thermal signature and a different surface texture that was already beginning to show fine cracking along its upper face as the clay dried and contracted in the morning air.
A targeted displacement of this section — breaking the looser clay matrix loose into the current — would open a second flow channel through the center of the obstruction. With two flow paths instead of one, the total channel cross-section available for water movement would increase significantly, reducing the upstream water level and slowing the rate of silt dissolution from the remaining dam material.
It would not eliminate the silt problem completely, but it would reduce the silt loading reaching the basin to a level that the ecosystem could absorb without catastrophic damage.
The question was how to break the loose section loose. Zephyr could not do it himself — his body mass was measured in milligrams, and the clay mass he was looking at weighed many kilograms. But displacement did not require direct physical force. It required understanding where force should be applied, and then assembling enough coordinated effort to apply it at the right point.
He completed his survey and memorized the exact location of the target section using the position of specific gravel stones on the exposed eastern channel bed as reference markers. Then he flew back downstream at full speed, following the river corridor below the willow canopy, the brown water of the silt plume tracking alongside him in the channel below.
The water strider colony was already in a defensive formation when Zephyr arrived at the basin — forty-three individuals clustered in a tight group near the central lily pad, their long legs spread on the water surface and their bodies oriented upstream, reading the surface vibrations of the incoming silt-laden current. The first brown tendrils of discolored water had reached the basin inlet and were spreading slowly across the surface in irregular ribbons.
Zephyr landed on the stem of a tall reed at the basin's northern edge and began his communication.
The texture of the dam's exposed surface provided additional diagnostic information that Zephyr gathered on his third survey pass. The clay face had two distinct zones visible to his compound eyes' ultraviolet-enhanced perception: a darker, moisture-saturated zone at the water interface where the clay was actively releasing particles, and a lighter, partially dried zone above the waterline where the surface had begun to crack in the morning air.
The crack pattern in the upper zone ran in irregular branching lines that were characteristic of material with uneven internal structure — exactly the loose, root-interlaced matrix he had identified as the primary target.
He hovered above the target section for three full minutes on his final assessment pass, reading every available signal from the structure below him. The air pressure differential was most pronounced at a specific point approximately forty centimeters from the dam's left edge — a location where a cluster of willow roots had created a particularly porous section in the clay.
At this point, the pressure differential reversed at certain wind angles, indicating that the section was already partially hollow behind the face — an air pocket or water-filled void that was absorbing the current's pressure rather than transmitting it through the clay mass.
This was the structural weakness that made the operation viable. A section with a void behind it would fail at lower applied force than the surrounding material. The combined pressure of the water striders' surface resonance, the diving beetles' current redirection structure, and his own air-vortex compression would be concentrated at this point, and the void would provide the initial movement that the rest of the operation needed to propagate.
He turned downstream and flew toward Pebble Bend at his best sustained speed.