One of the arguments presented in the article is that the program would alter the gene pool by 1% only. I'm curious where the 1% comes from and why something th

One of the arguments presented in the article is that the program would alter the gene pool by 1% only. I'm curious where the 1% comes from and why something that might feel "small" in another context should also apply here. What if the tolerance limit was 0.01%? How do we determine this gene pool variability tolerance?