Could someone with knowledge of birds and/or the methodology explain why the confidence intervals for the coasts and Arctic tundra biomes are so wide? Are thos

Could someone with knowledge of birds and/or the methodology explain why the confidence intervals for the coasts and Arctic tundra biomes are so wide? Are those bird populations just that difficult to count or estimate, is the data set just insanely small (if so, why?), or is there some other reason?