We investigated effects of regulated hunting on a puma (Puma concolor) population on the Uncompahgre Plateau (UPSA) in southwestern Colorado, USA. We examined the hypothesis that an annual harvest rate averaging 15% of the estimated number of independent individuals using the study area would result in a stable or increasing abundance of independent pumas. We predicted hunting mortality would be compensated by 1) a reduction in other causes of mortality, thus overall survival would stay the same or increase; 2) increased reproduction rates; or 3) increased recruitment of young animals. The study occurred over 10 years (2004–2014) and was designed with a reference period (years 1–5; i.e., RY1–RY5) without puma hunting and a treatment period (years 6–10; i.e., TY1–TY5) with hunting. We captured and marked pumas on the UPSA and monitored them year‐round to examine their demographics, reproduction, and movements. We estimated abundance of independent animals using the UPSA each winter during the Colorado hunting season from reference year 2 (RY2) to treatment year 5 (TY5) using the Lincoln‐Petersen method. In addition, we surveyed hunters to investigate how their behavior influenced harvest and the population. We captured and marked 110 and 116 unique pumas in the reference and treatment periods, respectively, during 440 total capture events. Those animals produced known‐fate data for 75 adults, 75 subadults, and 118 cubs, which we used to estimate sex‐ and life stage‐specific survival rates. In the reference period, independent pumas more than doubled in abundance and exhibited high survival. Natural mortality was the major cause of death to independent individuals, followed by other human causes (e.g., vehicle strikes, depredation control). In the treatment period, hunters killed 35 independent pumas and captured and released 30 others on the UPSA. Abundance of independent pumas using the UPSA declined 35% after 4 years of hunting with harvest rates averaging 15% annually. Harvest rates at the population scale, including marked independent pumas with home ranges exclusively on the UPSA, overlapping the UPSA, and on adjacent management units were higher, averaging 22% annually in the same 4 years leading to the population decline. Adult females comprised 21% of the total harvest. The top‐ranked model explaining variation in adult survival () indicated a period effect interacting with sex. Annual adult male survival was higher in the reference period ( = 0.96, 95% CI = 0.75–0.99) than in the treatment period ( = 0.40, 95% CI = 0.22–0.57). Annual adult female survival was 0.86 (95% CI = 0.72–0.94) in the reference period and 0.74 (95% CI = 0.63–0.82) in the treatment period. The top subadult model showed that female subadult survival was constant across the reference and treatment periods ( = 0.68, 95% CI = 0.43–0.84), whereas survival of subadult males exhibited the same trend as that of adult males: higher in the reference period ( = 0.92, 95% CI = 0.57–0.99) and lower in the treatment period ( = 0.43, 95% CI = 0.25–0.60). Cub survival was best explained by fates of mothers when cubs were dependent (mother alive = 0.51, 95% CI = 0.35–0.66; mother died = 0.14, 95% CI = 0.03–0.34). The age distribution for independent pumas skewed younger in the treatment period. Adult males were most affected by harvest; their abundance declined by 59% after 3 hunting seasons and we did not detect any males >6 years old after 2 hunting seasons. Pumas born on the UPSA that survived to subadult stage exhibited both philopatry and dispersal. Local recruitment and immigration contributed to positive growth in the reference period, but recruitment did not compensate for the losses of adult males and partially compensated for losses of adult females in the treatment period. Average birth intervals were similar in the reference and treatment periods (reference period = 18.3 months, 95% CI = 15.5–21.1; treatment period = 19.4 months, 95% CI = 16.2–22.6), but litter sizes (reference period = 2.8, 95% CI = 2.4–3.1; treatment period = 2.4, 95% CI = 2.0–2.8) and parturition rates (reference period = 0.63, 95% CI = 0.49–0.75; treatment period = 0.48, 95% CI = 0.37–0.59) declined slightly in the treatment period. Successful hunters used dogs, selected primarily males, and harvested pumas in 1–2 days (median). We found that an annual harvest rate at the population scale averaging 22% of the independent pumas over 4 years and with >20% adult females in the total harvest greatly reduced abundance. At this scale, annual mortality rates of independent animals from hunting averaged 6.3 times greater than from all other human causes and 4.6 times greater than from all natural causes during the population decline. Hunting deaths were largely additive and reproduction and recruitment did not compensate for this mortality source. Hunters generally selected male pumas, resulting in a decline in their survival and abundance, and the age structure of the population. We recommend that regulated hunting in a source‐sink structure be used to conserve puma populations, provide sustainable hunting opportunities, and address puma‐human conflicts. © 2021 The Wildlife Society.