If you're measuring an environment using a quadrat or another sampling instrument - why can't you just do it where you see the most biodiversity?

Taking a sample from an area of space where you can clearly see there to be more species richness or abundance is not taking a representative sample, and it won't tell you what is really going on in that particular space. It's important to make sure we get these estimations right because often we use them to assess how 'healthy' the environment is or how well a management stategy is doing, for instance, measuring indicator species that are sensitive to environmental conditions e.g. lichens sensitive to heavy metals can give us a good insight to how much pollution is present in an area, but if you intentionally capture more species then you could obtain a result that tells you that pollution isn't as bad as it actually is. So you need to make sure that the measurements are standardised e.g using a quadrat every 1 metre along a particular plane of environment, or kick sampling in a river to look for benthic macroinvertebrates for exactly 30 seconds in each identified location (e.g. every 1 metre upstream). It's also no good to just take a few samples, you need to repeat this process numerous times to make sure that your findings are reproducible and to ensure that anything you have found represents the environmental condition. This gives us a better idea about the biodiversity present in a location, it provides us with a more representative sample. Such estimates still have room for error, but by standardising our samples, we are hopefully minimising the opportunity for error.