Captive African (Loxodonta africana) and Asian (Elephas maximus) elephants can experience foot pathologies and arthritis. As a preventative measure against these pathologies and to alleviate the potential discomfort due to concrete substrates, some zoological institutions have renovated elephant housing to increase the amount of natural or shock-absorbent substrates. The objective of this study was to compare behavioral (diurnal and nocturnal) and glucorticoid (e.g., serum cortisol) responses of three female African elephants before, during, and after renovation to their indoor housing floor to assess whether renovations had short-term effects on the elephants' behavior and stress physiology. Behavioral data were collected using scan-sampling methods, and activity budgets were constructed for each of the three elephants. In addition, the duration of all lying rest activities were recorded. Weekly serum cortisol concentrations were determined with enzyme immunoassay (EIA). Overall, eating was the most prevalent behavior exhibited outdoors during the day, while resting (either in a lying or standing position) were most common during the indoor, nocturnal periods. Although variation existed among the three elephants, all three females spent significantly more time walking and less time eating during the day after the completion of the renovations. The extent to which the three elephants exhibited nocturnal lying rest behavior varied among the elephants, with the oldest elephant exhibiting the least amount (an average of 13.2 ± 2.8% of the nightly behavioral scans) compared to the two younger elephants (an average of 34.5 ± 2.1% and 56.6 ± 2.8% of the nightly behavioral scans). There was a significant increase in lying rest behavior for one elephant and standing rest for a second elephant following renovations. Baseline cortisol concentrations prior to renovations were 3.0 ± 0.4 ng/ml, 4.5 ± 0.5 ng/ml, and 4.9 ± 0.5 ng/ml for the three elephants. Cortisol concentrations remained baseline for two of the elephants throughout and after the renovation period, while one elephant that was pregnant had elevated cortisol during construction. Cortisol concentrations for the pregnant elephant remained higher than baseline once she was introduced to the new flooring and allowed back into the building, but these values were closer to the cortisol concentrations before renovations than during construction. Our findings demonstrate that individual elephants can vary in their behavioral and physiological responses to exhibit modifications. Given that the elephants walked more during the day, two of the three elephants had an increase in rest behavior during the night, and there were minimal changes in cortisol response after the flooring renovations, we conclude that the flooring renovations overall had a positive impact on animal welfare.