Validation of GPS-derived long-term trend in the atmospheric water vapor using homogenized radiosonde data

Conference poster, 2012

Atmospheric water vapor is important for the Earth’s energy balance due to its ability of absorbing and trapping long wave radiation emitted from the Earth’s surface. Therefore, measurements of the atmospheric water vapor content are of very interest for meteorology and climatology. Based on the path delay, which radio signals undergo when propagating through the neutral atmosphere, the Global Positioning System (GPS) measurements can be used to derive the atmospheric integrated water vapor (IWV). With a relatively high temporal resolution, continuously improved spatial density, and less expensive receivers, ground-based GPS measurement has been identified as a useful technique to monitor long-term variations in the IWV. This study will focus on investigating uncertainties of long-term IWV trend obtained from the NCAR global, 2-hourly ground-based GPS IWV dataset. The IWV trends were estimated for approximate 70 GPS sites covering the time period from 1997 to 2011. The GPS-IWV trends will be validated against that from the co-located and homogenized NCAR radiosonde data. The estimated GPS IWV trend consists of two types of uncertainties. The first type is larger in magnitude and is caused by short-term IWV variations (natural variability of the weather), which is not described by the model used for the trend estimation. This type will be investigated by using a statistical model and comparing the GPS IWV trends to the “true” IWV trends estimated from the radiosonde data during 1973-2011. The second type has smaller amplitude and is mainly caused by systematic errors in the GPS data, such as elevation-angle-dependent errors and changes of the GPS antenna and mask. A comparison with the homogenized radiosonde data will help us understand and quantify this type of uncertainty.