The definition of science by Richard Feynman.
A proton is a proton; you can make millions of 'em, and to within measurement accuracy of your equipment (and ignoring quantum effects), deflect them in an electromagnetic field exactly the same way. Einstein's general relativity was testable in Feynman's sense with a single observation: precession in the orbit of Mercury. Now take a look at the effect of diet on lipid metabolism. Even if you specified macronutrient ratios exactly, diets can vary tremendously from location to location, time to time, person to person. The system you're perturbing is composed of hundreds of individual chemical reactions linked in highly complex ways. And, each of us contains a mix of genes affecting those reaction rates in different ways. Therefore, advancing Nutrition Science is harder, because of the innate heterogeneity and complexity in the system, the inability to control the experimental/observational conditions precisely, and the relatively small number of experimental units.
Ecology is more like nutrition science than physics, in this regard.
Nutrition science also attracts greater interest from profit-making ventures in big ag and pharma than physics does; these interests can and do distort the scientific process in their own interests.
Ecology attracts it's own sort of value seeking interests; not usually profit motivated, but distorting the science in similar ways.
*ht to Peter Dunn on the Fat Head Facebook Group. This post is my initial comment and more explanation of that comment.