Water vapor mixing ratio and temperature inter-comparison results in the framework of the Hydrological Cycle in the Mediterranean Experiment\u2014Special Observation Period 1

Abstract

This paper reports results from an inter-comparison effort involving water vapor and temperature sensors, which took place in the North-Western Mediterranean in the period September–November 2012 in the framework of the first Special Observing Period of the Hydrological cycle in the Mediterranean Experiment. The involved sensors are the ground-based Raman lidars BASIL and WALI, the airborne water vapor differential absorption lidar LEANDRE 2, flying onboard the ATR42 aircraft, as well as additional water vapor and temperature sensors (radiosondes, aircraft in situ sensors, and a microwave radiometer). The main objective of the inter-comparison is the determination of the measurement uncertainty affecting these sensors. The effort benefitted from dedicated ATR42 flights in the framework of the EUropean Facility for Airborne Research (EUFAR) Project “WaLiTemp.” Comparisons between BASIL and LEANDRE 2 in terms of water vapor mixing ratio indicate a vertically averaged mean bias, overline{mathrm{bias}} , and mean absolute bias, left|overline{mathrm{bias}}right| , between the two sensors of − 0.08 g kg−1 (or − 2.50%) and 0.67 g kg−1 (or 2.77%), respectively. For all sensors’ pairs including LEANDRE 2, the inter-comparison range is 0.5–3 km, while for all other sensors’ pairs, the inter-comparison range is 0.5–6 km. Comparisons between BASIL and the microwave radiometer indicate overline{mathrm{bias}} and left|overline{mathrm{bias}}right| values between the two sensors of − 0.02 g kg−1 (or − 1.11%) and 0.22 g kg−1 (or 7.31%), respectively, for water vapor mixing ratio measurements, and a value for both overline{mathrm{bias}} and left|overline{mathrm{bias}}right| of 0.62 K for temperature measurements. Comparisons of BASIL with the radiosondes indicate overline{mathrm{bias}} and left|overline{mathrm{bias}}right| values of 0.28 g kg−1 (or 1.56%) and 0.51 g kg−1 (or 6.66%), respectively, for water vapor mixing ratio measurements, and − 0.43 K and 0.77 K, respectively, for temperature measurements, while comparisons of BASIL with aircraft in situ sensors indicate overline{mathrm{bias}} and left|overline{mathrm{bias}}right| values of 0.22 g kg−1 (or 1.17%) and 0.43 g kg−1 (or 4.62%), respectively, for water vapor mixing ratio measurements, and 0.15 K and 0.47 K, respectively, for temperature measurements. Comparisons of LEANDRE 2 with the radiosondes result in overline{mathrm{bias}} and left|overline{mathrm{bias}}right| values of 0.21 g kg−1 (or 0.76%) and 1.10 g kg−1 (or 11.05%), respectively, while comparisons of LEANDRE 2 with the aircraft in situ sensors indicate a value of both overline{mathrm{bias}} and left|overline{mathrm{bias}}right| of 0.76 g kg−1 (or 8.9%). Comparisons of in situ sensors with the radiosondes reveal overline{mathrm{bias}} and left|overline{mathrm{bias}}right| values of 0.36 g kg−1 (or 2.26%) and 0.36 g kg−1 (or 4.72%), respectively, for water vapor mixing ratio measurements, and 0.61 K and 0.62 K, respectively, for temperature measurements, while comparisons of in situ sensors with the microwave radiometer indicate overline{mathrm{bias}} and left|overline{mathrm{bias}}right| values of − 0.37 g kg−1 (or − 2.01%) and 0.58 g kg−1 (or 12.59%), respectively, for water vapor mixing ratio measurements, and a value of both overline{mathrm{bias}} and left|overline{mathrm{bias}}right| of 0.75 K for temperature measurements. Comparisons of the microwave radiometer with the radiosondes indicate overline{mathrm{bias}} and left|overline{mathrm{bias}}right| values of 0.13 g kg−1 (or 0.83%) and 0.20 g kg−1 (or 17.75%), respectively, for water vapor mixing ratio, and a value of both overline{mathrm{bias}} and left|overline{mathrm{bias}}right| of 0.74 K for temperature measurements. Our attention was also focused on the water vapor inter-comparison between BASIL and the transportable ground-based Raman lidar WALI, which took place in Candillargues on 30 October 2012, with results indicating overline{mathrm{bias}} and left|overline{mathrm{bias}}right| values between the two sensors of − 0.005 g kg−1 (or − 1.31%) and 0.24 g kg−1 (or 4.32%), respectively. Based on the available dataset and benefiting from the circumstance that BASIL could be compared with all other sensors, the overall bias of all sensors was also estimated. For water vapor mixing ratio measurements, the overall bias is 0.0079 g kg−1, 0.159 g kg−1, 0.084 g kg−1, − 0.201 g kg−1, 0.099 g kg−1, and − 0.141 g kg−1 for BASIL, LEANDRE 2, WALI, the radiosondes, the microwave radiometer, and the aircraft in situ sensor, respectively, being within ± 0.02 g kg−1 for all water vapor sensors. For temperature measurements, the overall bias is 0.11 K, 0.54 K, − 0.04 K, and − 0.51 K for BASIL, the radiosondes, the microwave radiometer, and the aircraft in situ sensor, respectively.