Abstract
Healthy coastal sabkhas (sabkha is an Arabic term for a salt flat) offer plenty of ecosystem services including climate change mitigation. However, fewer research studies were conducted at coastal sabkhas compared to other coastal marshes. This study was conducted in a total of ten coastal sabkha sites with different vegetation covers along the southern Red Sea coast of Saudi Arabia. The main objectives were to model and predict the distribution of volumetric soil organic carbon (SOC) density (kg C/m(3)) and cumulative SOC stocks (kg C/m(2)) using three different mathematic functions (allometric, exponential, and sigmoid) based on sampled and observed soil carbon (C) data (total of 125 soil cores = 1250 soil samples). Sigmoid function showed the greatest fit for predicting the distribution of volumetric SOC density over soil profile depth with mean Adj. R-2 = 0.9978, 0.9611, and 0.9623 for vegetation cover of >25-50, >50-75, and >75-100%, respectively. For modeling the cumulative SOC stocks, both validation indices and p of the t-test confirmed that using the exponential function is the most appropriate to be used for predicting the SOC stock among different vegetation covers. Moreover, assessing the topsoil concentration factors (TCFs) showed that the distribution of the SOC content is impacted to a great extent by the vegetation cover at coastal sabkhas. Sampling the soil parameter of interest to estimate the SOC stocks is constrained by time and cost. Therefore, using the exponential function for predicting the distribution of cumulative SOC stocks at coastal sabkhas over soil profile depth is appropriate and promising for mapping SOC stocks at both regional and global spatial scales.