What I am going to say is the most classical approach to quantify feeding on a resource by animals.

So, per capita ingestion rates should depend on two factors:

1. The species of larval fish, which determines, for instance, an intrinsic attack rate

2. The density of a resource, for instance, zooplancton, that larval fish feeds on.  Per capita ingestion rates should increase with resource availability.  However, it is reasonable to assume also that, when resource availability is very high, per capita ingestion rates should level off and reach some asymptotic value.

These two assumptions brings us to the famous Holling II type functional response. This is a two parametric function that has been fitted to data quite often since Holling paper:

C. S. Holling (1965). The Functional Response of Predators to Prey Density and its Role in Mimicry and Population Regulation. Memoirs of the Entomological Society of Canada, 97, pp 5-60. doi:10.4039/entm9745fv.

Since I am not an expert in larval fish feeding, my answer may sound very basic to you. However, let me say that whatever we try to do, it is very nice and fair to look always for connections to classical work. If this answer ends up not being very useful to you, it will be at least an acknowledgement to one of the classics in Ecology.

This manuscript (attached) is a nice review of the relevant literature.

MacKenzie, B. R., W. C. Leggett, and R. H. Peters. "Estimating larval fish ingestion rates: Can laboratory derived values be reliably extrapolated to the wild?." Marine ecology progress series. Oldendorf 67.3 (1990): 209-225.

However, don't forget that ingestion is also related to turbulence. For example:

MacKenzie, Brian R., et al. "Evidence for a dome‐shaped relationship between turbulence and larval fish ingestion rates." Limnology and Oceanography 39.8 (1994): 1790-1799.

Kiorboe, Thomas. "Turbulence-enhanced prey encounter rates in larval fish: effects of." Journal of Plankton Research 17.12 (1995): 2319-2331.

Good Luck!

Shannon