The Passive Microwave Water Cycle (PMWC) Product: Closing the Water Cycle Using a Constellation of Satellites
Abstract
We have developed a water cycle product as part of the NASA Energy and Water Cycle Study (NEWS). The purpose of the product is to integrate passive microwave retrievals from a variety of different sensors on different satellites including SSMI (F08, F10, F11, F13, F14, and F15), SSMIS (F16 and F17), AMSR (Aqua and Midori-II), TMI on TRMM, WindSat, and eventually AMSU (NOAA-15 and NOAA-16). The water cycle over a particular location averaged over a time scale of one month is given by: E-P=WVTD; where E is evaporation, P is precipitation, and WVTD is water vapor transport divergence. The new and unique feature of our product is that we make use of the large quantity of accurately intercalibrated water vapor and wind observations in order to estimate WVTD. As part of developing this product we have evaluated our new intercalibrated microwave rain rates, developed a procedure for estimating evaporation, and developed a procedure for estimating water vapor transport and its divergence. The Version-01 Passive Microwave Water Cycle (PMWC) dataset will contain maps of evaporation, precipitation, water vapor transport, water vapor transport divergence, and water vapor. Uncertainty estimates for each parameter will also be supplied. Currently, the product is a 20-year (1987-2007), 0.25-degree, monthly average product over the global oceans. One of our principle motivations is to obtain estimates of the uncertainty in "direct" physically-based retrievals of precipitation. Direct physically-based rain retrievals are subject to large uncertainties that are hard to quantify, such as horizontal inhomogeneity (beamfilling), cloud and rain water partitioning, rain column height and the rain vertical profile, drop size distribution, and the effects of frozen hydrometeors. By using the balanced water cycle, we can estimate precipitation uncertainties in P by estimating uncertainties in E and WVTD. Estimating uncertainties in E can be done with a straight-forward classical uncertainty analysis. Uncertainties in WVTD can easily be estimated using on-orbit simulation experiments with model data. In addition to uncertainty estimates, we also note that estimating precipitation through balancing the water cycle provides a new and independent estimate of precipitation. The errors with this technique are independent from the errors in passive microwave, active microwave, and infrared precipitation retrievals. This will be of great value as we enter the Global Precipitation Measurement (GPM) era. Moreover, consistency among hydrological parameters (evaporation, precipitation, water vapor transport divergence) and especially their trends provides indirect validation of the retrievals. Currently, we have evaluated trends in our evaporation, precipitation, and water vapor datasets on a global average basis. We find water vapor trends of 7 Percent/degree as the world warms - consistent with the Clausius-Clapeyron (C-C) relationship and with climate models. We also find that our evaporation and precipitation measurements both increase at close to the C-C rate. This is in contrast with climate models that predict a muted response of precipitation to global warming with rates between 1 to 3 Percent/degree. These water cycle relationships will be discussed in terms of their implications for the global energy balance.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2007
- Bibcode:
- 2007AGUFM.H33I..07H
- Keywords:
-
- 1640 Remote sensing (1855);
- 1655 Water cycles (1836);
- 1836 Hydrological cycles and budgets (1218;
- 1655);
- 1854 Precipitation (3354);
- 1855 Remote sensing (1640)