SPONSOR:
National Institute for Water Resources, Water Resources Research Institute Program
PROJECT PERIOD:
3/1/2015 - 2/28/2016
ABSTRACT:
Understanding of the amount, spatial distribution, and temporal variability of evapotranspiration (ET) in Hawaii is essential for assessment of water resource availability and management of water resources. The accurate estimation of water flux between the land surface and atmosphere plays an important role in many fields such as irrigation, estimation of potential recharge, impacts of climate change on stream flows and groundwater, restoration of ecosystems, etc., in Hawaii. Giambelluca et al. (2014) used the Penman-Monteith equation to estimate ET in Hawaii. This approach needs the a priori specification of aerodynamic and stomatal resistances as well as ground heat flux. This is difficult because it is based on empirical and site-specific relations with a large uncertainty. This can cause large errors in ET estimates. Bateni et al. (2013a) overcame the such shortcomings by developing a new ET retrieval model that does not require any empirical or site-specific equations. Their approach estimated ET by assimilating sequence of land surface temperature, LST (as the state variable of the land surface) observations into a variational data assimilation (VDA) framework. This model is promising for ET retrieval at dry/sparsely vegetated sites but weak for wet/densely vegetated land surfaces. The main purpose of this proposed research is to build a new VDA model that provides accurate ET estimates in both dry/lightly vegetated and wet/heavily vegetated sites. The model also allows us to evaluate the impact of climate change on ET in future studies. The new approach builds on the VDA method introduced by Bateni et al. (2013a), but it advances that approach in two major new directions: First, it takes advantage of information in sequences of not only LST (as the state variable of land surface) but also reference-level air temperature and specific humidity (as the state variables of the lower atmosphere). Second, it assimilates LST, air temperature and specific humidity into a “coupled” land surface-atmospheric boundary layer (ABL) model. Such a coupling is advantageous because it allows for interaction between the land surface and overlying atmosphere. The new model will be used to estimate ET over the Montane Cloud Forests (MCFs) in Hawaii and several sites with different hydrological conditions in the mainland. It is expected that it provides accurate ET estimates in various sites sampling a wide variety of hydrological conditions. This project generates a robust ET retrieval model and preliminary results for our upcoming proposal which aims at 1) retrieving ET over the whole state of Hawaii and 2) evaluating the impact of climate change on ET in Hawaii.
PRINCIPAL INVESTIGATOR
