The surface and leaf temperatures in various places during the day in Puerto Rico in summer

Despite the widely recognized greenhouse effect by greenhouse gases, atmospheric temperature is also directly driven by ground surface temperature. Tropical forests cover 7% of terrestrial land and likely play a noticeable and disproportionable role in regulating the temperature of the ground surface that emits infrared radiation available for the subsequent greenhouse effect. Here, we evaluated the role of plant leaves in modulating daytime ground surface temperature across tropical dry-rain forests in Puerto Rico. We found that leaf midday temperature decreased with altitude following a logistic function. In contrast, bare ground surface temperature decreased linearly with altitude. Canopy cooling effect ranged from 34.5℃ to -10.2℃, with the strongest cooling observed in low-altitude forests and diminishing at higher elevations. Further, dicotyledons exhibited a stronger cooling effect on ground surface temperature than pteridophytes and monocotyledons. Dry forests showed the greatest cooling effect at high temperatures, whereas wet-rain forests exhibited the most effective cooling when temperatures were below 10℃. These findings highlight the critical yet overlooked role of tropical forests in regulating ground daytime surface temperature and call for greater attention to the biophysical impacts of deforestation on climate warming beyond the carbon-centric discourse.Samples were collected from plants growing in seven regions in the Commonwealth of Puerto Rico (Table 1) between 10 am and 4 pm. We measured the longitude and latitude of every region via the handheld GPS terminal and the data are presented in World Geodetic System 1984 (WGS84). We categorized all the measured regions into four groups based on wide life zones in the tropics. The dry forest receives 600-1100 millimeters of rainfall per year, moist forest receives 1000-2200 millimeters of annual rainfall, and wet-rain forests receive more than 2000 millimeters of annual rainfall. Elevation data, and  were measured at the same time as the samples were taken. Hence, for each site, those data may vary, depending on the specific location of the sampling point and the weather condition of the sampling time and the weather condition of the sampling time. Previous studies have shown that under field conditions, within a certain wind speed range (0 to 2 m/s), changes in wind speed have almost no effect on leaf temperature, but for the sake of seriousness, we still got wind speed data from ERA5. We collected samples from 67 tropical species (Table S1) from lowlands up to montane-forest including monocots (monocotyledonous plants), dicots (dicotyledonous plants), and ferns (pteridophytes). In order to choose the species that are beneficial to the research, we have followed the principle of preferential selection to include widespread species such as Hedychium coronarium (White ginger lily), Cyathea arborea (West Indian tree-fern), and Prestoea acuminata var. montana (Sierra palm).41 Species included primarily trees, but also woody shrubs and herbaceous plants.At each sampling site, the sun-exposed outer canopy leaves were selected for measurement. Our measuring time was from 2 p.m. to 4 p.m. each day to ensure stable lighting conditions. The reachable leaves were measured directly with the infrared thermometer (Etekcity Infrared Thermometer 774) on the top side of the leaf surface. The leaves at the high canopy were cut using pole pruners and immediately measured within five seconds. For each plant, we randomly chose at least five different leaves that fit the criteria and measured them separately to obtain an average to improve the reliability and repeatability of the data.