H21J-01 INVITED
Large Scale Field Campaign Contributions to Soil Moisture Remote Sensing
Large-scale field experiments have been an essential component of soil moisture remote sensing for over two decades. They have provided test beds for both the technology and science necessary to develop and refine satellite mission concepts. The high degree of spatial variability of soil moisture and the relatively coarse resolution of satellite observations present significant challenges for scaling and the design of these field campaigns. Earlier experiments, in particular Washita'92, established the credibility of large scale application of microwave remote sensing. Of particular significance was the demonstration of synthetic aperture radiometry (now the core technology of the Soil Moisture Ocean Salinity, SMOS, mission), spatial mapping and consistent retrievals, and the temporal information content of repeat observations in deriving soil hydraulic properties. Basic concepts were expanded in both space and time in experiments such as SGP97 that attempted to integrate the soil moisture observations in the broader framework of land surface hydrology. These types of campaigns have expanded globally. Within the U.S. in recent years they have focused on two issues; establishing the foundations of a future active-passive soil moisture satellite mission and the development and validation of current satellite soil moisture algorithms. Experiments involving a range of spatial domains (point, field, region) and geographical domains have contributed to establishing scaling relationships for both measurements and processes. Future experiments must continue to address the needs of planned soil moisture missions such as SMOS and Aquarius. As the Soil Moisture Active Passive (SMAP) mission concept continues to develop, there will be an increasing need to test and refine algorithms and in the post-launch time frame, there will be a need for validation products. These needs must be integrated with broader science objectives (to the degree it is possible) in order to secure the required resources. Efforts should be made to find communality with other missions and other research programs.
H21J-02
Microwave, Water and Energy Balance Experiments (MicroWEXs) for multidisciplinary studies in remote sensing, hydrology, and agriculture
The need for long-term remote sensing observations in different terrain and climatic regions for development/validation of hydrologic models and assimilation methodologies cannot be over-emphasized. Observations at microwave wavelengths are particularly sensitive to near-surface soil moisture. Many studies have been conducted in agricultural terrain such as bare soil, grass, soybean, wheat, pasture, and corn to understand the relationship between soil moisture and microwave remote sensing. However, most of these experiments have been short-term experiments capturing only a part of growing seasons. Microwave, Water, and Energy Balance Experiments (MicroWEXs) are a series of season-long experiments conducted at a 9-acre field site to understand land-atmosphere interactions and microwave observations for growing corn and cotton in North Central Florida. During MicroWEXs, we observed microwave signatures at 4.5 cm (MicroWEX-1through 6) and 20 cm (MicroWEX-5 and 6) using tower mounted radiometer systems, along with co-located concurrent micrometeorological parameters. In addition, we observed detailed soil moisture, temperature, and heat flux profiles at multiple locations, ground water levels, and soil and vegetation parameters throughout the growing seasons. Such extensive, high temporal frequency datasets are being used for understanding physically-based SVAT, crop growth, and microwave remote sensing models and developing novel assimilation technologies to improve model estimates of moisture and energy fluxes at the land surface and in the vadose zone. An overview of various observations during the MicroWEXs and their significance in multidisciplinary studies will be presented.
H21J-03 INVITED
Evaluation of High Resolution Precipitation Products in Northwest Mexico during the North American Monsoon Experiment (NAME)
This study examines the spatial and temporal variability of the diurnal cycle of clouds and precipitation tied to topography within the North American Monsoon Experiment (NAME) Tier-1 domain during the 2004 NAME Enhanced Observing Period (EOP, July-August). We focus on the implications for high resolution precipitation estimation and hydrologic studies within the core of the monsoon. Ground-based precipitation retrievals from the NAME Event Rain gauge Network (NERN) and CSU/NCAR version 2 radar composites over the southern NAME Tier-I domain are compared with satellite rainfall estimates from the CMORPH, TRMM 3B42, and PERSIANN operational satellite estimates to examine timing and magnitude differences in their representation of the diurnal cycle along the western slopes of the Sierra Madre Occidental (SMO). Gauge and radar data are examined alongside hourly images of high resolution 11-μm brightness temperature from GOES and cloud to ground lightning flash rates from the North American Lightning Detection Network to investigate diurnally- evolving cloud structures and inferred microphysics. Over high terrain, it is found that convection is relatively shallow (in terms of the depth of mixed-phase processes), although precipitation and lightning often occur near or just after noon local time. It is hypothesized that this shallow cloud and mixed-phase depth tends to contribute to an underestimation of precipitation from both IR and microwave precipitation algorithms. Once the convection has evolved or propagated into lower elevations (around 1500 LT) where more moisture is available, deep, tropopause-depth convection according to IR cloud top temperatures results producing a maximum of lightning and rainfall. Thereafter, organized, deep convection at times propagates onto the coastal plain in the form of mesoscale convective systems where it typically dissipates, although longer lasting deep convection is occasionally observed to propagate out across the Gulf of California. At lower elevations, high sub-cloud evaporation rates and cirrus shielding appear to be contributing to a relative overestimate of precipitation by non- gauge corrected microwave and infrared-based satellite rainfall algorithms; radar estimates showed better correspondence with gauges, presumably due to a lower radar beam height above ground. The remotely sensed precipitation products were then used to drive an operational land surface model in order to assess their relative impacts on runoff, soil moisture and surface energy flux partitioning. The impact of the error structures of key precipitation characteristics (duration, intensity and frequency) from the various products on modeled surface hydrologic variables is found to be significant. Based on these analyses recommendations on future precipitation monitoring for the North American Monsoon system are made.
H21J-04
Evaporation and Transpiration in Semiarid Grass- and Shrub-Dominated Ecosystems in Southeast Arizona
Information about the ratio of transpiration (T) to total evapotranspiration (T/ET) is related to critical global change concerns, including shrub encroachment, non-native species invasion and soil erosion. In this study, a new approach was used to partition measurements of ET into daily evaporation (ED) and daily transpiration (TD) in a semiarid watershed based on the low-cost addition of an infrared thermometer and soil moisture sensors to existing eddy covariance and Bowen ratio systems. For a study period during the North American monsoon season, estimates of T/ET over three years (2004-2006) at grass- and shrub-dominated sites at the USDA Walnut Gulch Experimental Watershed were used to address the hypotheses that T/ET is sensitive to changes in woody plant cover and is correlated with total precipitation, precipitation patterns and total ET. For this study period (August to October), we found a strong, multi-year relation between root zone soil moisture (to 15 cm depth) and TD at both sites. Estimates of TD and ED were summed over the study period for years 2004, 2005 and 2006 to estimate totals over the study period, TS and ES respectively. Results showed that the shrub-dominated site had higher ES than the grass-dominated site for similar precipitation patterns over the study period. TS and ES were related to the number of larger storms during the study period, and this relation was different for the two sites. For this study period, TS was related strongly to ETS, with a slope of 0.90 for the grass-dominated site and 0.84 for the shrub-dominated site for the three years. Thus, for these sites during the study period in these years, the TS/ETS was higher for the grass-dominated site than for the shrub-dominated site, and did not vary systematically with variation in amounts and timing of rainfall. The uncertainty of the new method due to sampling, instrument and algorithm errors was estimated to be about 3-7 mm or about 4% of TS over the study period.
H21J-05 INVITED
A Top-Down Approach for Estimating Effective Soil Hydraulic Parameters from Space
The estimation of effective soil hydraulic parameters and their uncertainties is a critical step in all large-scale hydrologic and climatic model applications. In this study, a scale-dependent (top-down) parameter estimation (inverse modeling) scheme called the Noisy Monte Carlo Genetic Algorithm (NMCGA) was developed and tested for estimating these effective soil hydraulic parameters and their uncertainties. We tested our method using three case studies involving a synthetic pixel, an airborne remote sensing (RS) footprint, and a satellite RS footprint. In the synthetic case studies with pure (one soil texture) and mixed-pixel (multiple soil textures) conditions, we found that the NMCGA performed well in estimating the effective soil hydraulic parameters even with the complexities of various soil types and land management practices. Using airborne or satellite remote sensing soil moisture data, the NMCGA was found to be suitable for estimating the effective soil hydraulic properties that could mimic large-scale soil moisture time-series if used in forward stochastic simulation models. The results also showed that the effective soil water retention curve è(h) tend to scale down (smaller mean) at the larger satellite remote sensing pixel compared to air-borne remote sensing pixel. This finding, however, did not generally imply that every effective soil hydraulic parameters have to be scaled down like the soil water retention curve. The Mualem- van Genuchten soil hydraulic parameters á and n tend to increase (in mean and spread) as the parameter search spaces were relaxed progressively in our satellite remote sensing studies. The scaling down of the soil hydraulic parameters was observed to be more profound in èsat than that of the other scale parameters such as Ksat and ères. Overall, the NMCGA framework was found to be very promising in the inverse modeling of remotely sensed near-surface soil moisture for estimating the effective soil hydraulic parameters and their uncertainties at the remote sensing footprint/climate model grid.
H21J-06
The Determination of Soil Hydraulic Properties From Surface Temperature and the Effect of Errors in Other Model Parameters
Soil hydraulic properties (SHPs) are perhaps the least well constrained parameters in land surface models. Other land surface model parameters (e.g. vegetation type and cover, albedo, slope, and roughness) are observable at the land surface, and thus are amenable to measurement by satellite. However, SHPs have no clear representation at the surface, and thus are largely unknown. Present estimates of soil properties come from maps of soil texture, and have been shown to be extremely poor (r2=0.05). Previous work has suggested that calibrating SHPs using remotely sensed surface temperature has promise. Here, we show that using surface temperature to calibrate SHPs can dramatically improve model estimates of evapotranspiration. This process decreases mid-day RMS errors from 110 to 60W/m2 (4.2 to 2.3 mm/day), and decreases the bias from 30 to 2 W/m2 (from 1.1 to 0.076 mm/day). However, this requires that all other parameters in the model are reasonably well known. Errors in parameters such as surface roughness, or in input data such as solar radiation or air temperature can cause the SHP estimation process to increase errors in evapotranspiration in attempts to compensate. Errors in other evapotranspiration sources (e.g. precipitation, ground water availability, or vegetation cover) lead to errors in the estimated SHPs, but these errors offset each other, and the bias between modeled and measured evapotranspiration is reduced.
H21J-07
What is the ability of distributed hydrologic models to reproduce observed spatial soil moisture fields?
A common approach of evaluating distributed hydrologic models is to compare predicted and observed streamflows. Although the total runoff generated in a watershed is compared with observed sreamflow, it is often difficult to understand what is going on inside the watershed as many possible scenarios can result in the same integrated response. Spatiotemporal datasets from intensive hydrological field experiments offer opportunities to validate the spatial hydrologic fields estimated by distributed hydrologic models and tie the integrated basin response to the internal soil moisture field. In this study, we use spatial soil moisture observations gathered during the Southern Great Plains 1997 (SGP97) field experiment in the Little Washita watershed (central Oklahoma) to evaluate the accuracy of simulated soil moisture fields from a distributed hydrologic model known as tRIBS (TIN-based Real-time Integrated Basin Simulator). The soil moisture observations are from the aircraft- based ESTAR (Electronically Steered Thinned Array Radiometer) sensor as well as in-situ field observations. The results of this study will be useful for assessing the utility of distributed hydrologic models to examine the spatial properties of hydrologic fields.
H21J-08
Diagnosis of performance of the Noah LSM snow model
Quantifying snow water equivalent (SWE) and runoff timing are vital in hydrologic modeling, especially for snowmelt driven regimes such as the western U.S. and the pan-Arctic. The Noah land-surface model, used in the National Center for Environmental Prediction's operational suite of weather and climate forecast models, has shown a tendency toward ablation of snowpacks earlier than in observations in off-line tests. We report a diagnosis of the sensitivities of various components of the snow scheme in the Noah model that might affect snow ablation. We find that issues with the model's turbulent exchange schemes, and its radiative exchange formulations appear to be the main contributors to the negative SWE bias during the ablation season. Using modified albedo and surface exchange schemes, we test alternative formulations both using point observations at observing sites in mountain maritime locations in the Pacific Northwest, and in the interior of the western U.S. and Canada, as well as in simulations of seasonal variations of snow extent over the entire pan-Arctic drainage. Our results suggest that the a modest change in the snow albedo formulation results in the greatest improvement in model performance, although under some conditions an alternation in the turbulent exchange parameterization produces improved results as well.