N‐mixture models are often employed to estimate latent organismal abundance while concurrently accounting for detection probability. Our study offers a novel means for simultaneously measuring abundance, detection probability, and gear efficiency by focusing on a previously understudied and relatively immobile subject (river herring eggs). Custom‐designed egg mats accommodating two individual collecting surfaces were deployed in two tributaries of the Hudson River (Fall Kill and Black Creek, New York) to collect anadromous river herring eggs during the spawn. Mats were orientated approximately parallel to streamflow under a stratified random sampling design. Strata were defined as three equidistant spatial segments measured from a given tributary's confluence with the main stem of the Hudson River to its first impassable barrier to fish migration. In total, 93 sites were surveyed, with the majority of eggs being detected within the upper two‐thirds of each respective tributary. Throughout the course of the sampling season, an average of 1,585 eggs per egg mat subsampling event was recovered from the upper two strata of Black Creek, and 2,619 eggs per subsampling event were recovered from the upper two strata of the Fall Kill. In Black Creek, an egg density of 568 eggs per 58.1‐cm2 egg mat subsample was observed over an average deployment duration of 3.7 d at a detection rate of 82%. In the Fall Kill, an egg density of 1,222 eggs per 58.1‐cm2 subsample was observed over an average deployment duration of 3.9 d at a detection rate of 92%. The N‐mixture negative binomial model outperformed Poisson and zero‐inflated Poisson N‐mixture models in estimating river herring egg abundance using Akaike's information criterion model comparison indices. In terms of iteration processing time, N‐mixture modeling using the “unmarked” package in R proved to be more efficient than Bayesian‐based hierarchical modeling processed through the “jagsUI” package.