Retrieval of atmospheric and ocean surface parameters from ADEOS‐II Advanced Microwave Scanning Radiometer (AMSR) data: Comparison of errors of global and regional algorithms

Abstract

Retrieval of the sea surface temperature t s , wind speed W , total atmospheric water vapor content V , and total cloud liquid water content Q over the ocean from the simulated ADEOS‐II AMSR data in the absence of precipitation is considered. The brightness temperatures ( T B ) with the vertical (V) and horizontal (H) polarizations at the AMSR frequencies ν = 6.9, 10.7, 18.7, 23.8, and 36.5 GHz were computed for the radiosonde and relevant data collected by research vessels. V and Q were retrieved from T B (23.8V) and T B (36.5V) with the physically based “global” (−1.6 < t s ≤ 31°C), “polar” ( t s ≤ 15°C), and “tropical” ( t s ≥ 24°C) algorithms under the assumption that t s values were derived from the measurements at 6.9 and 10.7 GHz with an error σ ts . The errors σ V and σ Q were estimated at several combinations of the difference Δ T 36 = T B (36.5V) ‐ T B (36.5H), radiometer noises Δ T and σ ts . At Δ T 36 = 35 K, Δ T = 0.3 K, and σ ts = 1°C, σ V = 1.5 kg/m 2 and σ Q = 0.029 kg/m 2 for a global algorithm. Standard regression techniques were applied to retrieve t s and W from the simulated brightness temperatures for the cases T B (10.7V) ≤ 185 K. For a global data base, three‐channel algorithm (6.9V, 6.9H, 10.7V or 10.7V) yields the t s and W errors equal to 0.40 and 0.58°C and 0.66 and 0.85 m/s as radiometer noises increase from 0.1 to 0.2 K at ν = 6.9 GHz and from 0.13 to 0.27 K at ν = 10.7 GHz.