Evaluation of atmospheric water vapor product from the MODIS online visualization and analysis system during the summer of 2015

• Main Authors: Castilla, Red, Vallar, Edgar A., Galvez, Maria Cecilia, Cayetano, Mylene, Simpas, James, Cruz, Melliza, Lim, Rheo Lamorena, Lim, Len Herald, Pabroa, Preciosa Corazon, Macatangay, Ronald, Bagtasa, Gerry, Morris, Vernon
• Format: text
• Published: Animo Repository 2015
• Subjects: Water vapor, Atmospheric; Moisture; Physics
• Online Access: https://animorepository.dlsu.edu.ph/faculty_research/2007
• Institution: De La Salle University

Atmospheric water vapor plays a significant role in radiation balance and biogeochemical processes. As the primary greenhouse gas, its quantification is very essential in a wide range of application such as numerical weather prediction, radiative transfer modeling, regional climate modeling, and air quality studies. Columnar measurements of water vapor, or precipitable water (PW) are necessary in many of these applications and can be obtained by remote sensing and in-situ methods. Moderate Resolution Imaging Spectroradiometer (MODIS) onboard the polar-orbiting Terra and Aqua platforms, can monitor precipitable water with high accuracy and large scales over land and ocean, using near-infrared (nIR) bands on clear sky conditions. The MODIS nIR PW is only available during daytime, which can be retrieved at 1°×1° spatial resolution from the MODIS Online Visualization and Analysis System (MOVAS). MODIS water vapor product at the 931-941 nm bandwidth was evaluated by comparing with coincidental ground-based measurements of PW using a handheld spectral solar radiometer (Microtops II) measuring PW at 936 nm during a measurement campaign of the summer of 2015. Further, PW values derived from atmospheric soundings by the nearest radiosonde station were used to evaluate the accuracy of the different PW measurements. Systematic comparisons among the different platforms, i.e. bivariate correlation, absolute and relative root mean square error, absolute and relative mean biased error were performed. Multiple linear regression analysis to calibrate the MODIS PW from the nIR band using Microtops II and radiosonde observations within MODIS image coverage are presented and assessed.