6000至7000只是两类鸟的合计快照
来源称东平湖近期观测到野生鸬鹚和花脸鸭共6000至7000只，后续还会有候鸟抵达；其中有的短暂补给后继续南迁，有的在湖区越冬。报道没有给出两种各自的精确数量，因此不能把合计数分别套给鸬鹚或花脸鸭。
数字对应一次阶段性观察，而非某一观景台的每日保证。记录时应注明日期、观察时段、天气、水位与覆盖岸段，并把飞入、飞出和可能重复的群分开。
先按行为与轮廓分层，再估算密度
鸬鹚体型大、长颈长尾，常低贴水面成线飞行，停栖时可见张翼晾羽；花脸鸭更小、更紧凑，雄鸟脸部绿黄黑色图案醒目，雌鸟辨识更难，需要结合脸侧淡斑、嘴基小白点、体形和与同群个体的比例。
先把水面、空中和岸边停栖鸟分开，再以20只或50只为一个样本块，按相似密度分区累加。大群突然起飞时不要只数“黑点”，可拍连续广角画面，事后逐格核对并给出合理区间。
在淡水湖边把安全距离写进记录
只在公共岸线、堤路或观景设施停留，不开车下滩、不穿越养殖区、不乘船追群，也不使用无人机、鸣笛、投喂和回放。若鸟群整体游离岸边或连续惊飞，应停止移动并后退。
湖面风浪、雾和逆光会同时降低辨识与计数质量，应如实记录“未定鸭类”或数量区间。公开记录只保留湖区级范围，不发布夜栖岸段、稳定觅食点或鸟群实时移动方向。
Six to seven thousand is a combined snapshot
The source says 6,000–7,000 wild cormorants and Baikal Teal were recently observed at Dongping Lake, with more migrants expected. Some birds will refuel and continue south, while others may winter. It does not provide separate totals, so the combined number must not be assigned to either species alone.
The figure is a dated survey snapshot, not a daily guarantee at one viewpoint. Record date, observation period, weather, water level and shore coverage, and separate arrivals, departures and possible repeat groups.
Separate silhouette and behaviour before scaling density
Great Cormorants are large, long-necked and long-tailed, often fly in low lines and may hold wings open when perched. Baikal Teal are smaller and compact; the male has a striking green, yellow and black facial pattern, while females require a combination of the pale face spot, small white mark at the bill base, structure and comparison with flock mates.
Separate birds on water, in flight and on shore first. Use a sample block of twenty or fifty and scale across zones of similar density. When a large flock erupts, take sequential wide frames rather than counting generic dots, then review frames and report a defensible range.
Make distance part of a freshwater-lake record
Stay on public shores, embankments or visitor structures. Do not drive onto flats, enter aquaculture areas, pursue by boat, fly drones, honk, feed or use playback. If the flock swims away together or repeatedly flushes, stop moving and retreat.
Wind, haze and backlight reduce both identification and count quality. Record unidentified ducks or a numerical range honestly. Public records should remain at lake scale and omit roost shores, dependable feeding points and live movement directions.