There are many carbonate reservoirs in US (and the world) with light oil and fracture pressure below its minimum miscibility pressure (or reservoir may be naturally fractured). Many carbonate reservoirs are naturally fractured. Waterflooding is effective in fractured reservoirs, if the formation is water-wet. Many fractured carbonate reservoirs, however, are mixed-wet and recoveries with conventional methods are low (less than 10%). Thermal and miscible tertiary recovery techniques are not effective in these reservoirs. Surfactant flooding (or huff-n-puff) is the only hope, yet it was developed for sandstone reservoirs in the past. The goal of this research is to evaluate dilute (hence relatively inexpensive) surfactant methods for carbonate formations and identify conditions under which they can be effective. Laboratory imbibition tests show about 61% oil recovery in the case of Alf-38 and 37% in the case of DTAB. A numerical model has been developed that fits the rate of imbibition of the laboratory experiment. Field-scale fracture block simulation shows that as the fracture spacing increases, so does the time of recovery. Plans for the next quarter include simulation studies.