There are over 300 coastal islands in Texas that collectively provide nesting habitat for >250,000 colonial waterbirds.  Most of these islands were created as a by-product of dredging the Intracoastal Waterway, which was completed in 1949.  Since then, islands have been eroding, potentially limiting waterbird nesting populations.  Managers are responding by putting large sums of money into rehabilitating the islands.  However, there is uncertainty in how fast the amounts of different nesting substrate types (vegetated and unvegetated) are changing and in the factors that are affecting the rates of change.  Therefore, we digitized aerial imagery of 213 islands along the Texas coast south of Aransas Pass, Texas, for the years 1995, 2008, 2012, and 2018 to estimate changes in rookery island area and area of bare and vegetated substrates. During the 23 years of our study, 23 of the 213 study islands were completely submerged.  Overall, islands lost emergent substrate at a median annual rate of 2.2% (IQR = 4.8%). Islands lost bare substrate at a median annual rate of 3.0% (IQR = 2.0%) per year, and they lost vegetated substrate at a rate of 1.9% (IQR = 3.3%) per year. For a subset of 183 islands for which we had a complete set of covariates, we analyzed their effects on the rate of habitat changes.  Nesting habitat loss was highest for small islands, islands with low elevation, and those with a low percentage of vegetated cover. The location of an island in the landscape also affected rate of habitat loss, with islands located close to navigation channels and those with a long fetch to the northwest or southeast having the highest rates.  We were not surprised to find that bare substrate, typically sand and shell, was lost at a higher rate than was vegetated habitat because bare substrate tends to be an early successional stage of an island, often close to the waterline.  Further, we hypothesized that the bare substrate loss could be partially the result of vegetation succession; however, the net loss of vegetated substrate does not support this hypothesis.  If habitat loss continues at, or above, the observed rates, many islands will be submerged in the next few decades, underscoring the urgent need for rookery island restoration to sustain colonial waterbird populations in Texas.