1.0 The Science Imperative
2.0 Historical Context for a Soil Moisture Mission
2.1 Underlying Physics
2.2 Heritage of Large Antennas
2.3 Large Footprint Scale and Heterogeneity Issues
2.4 Linking Surface Moisture to the Profile Soil Moisture
3.0 Mission Rationale
4.0 Baseline Mission
5.0 Science Roadmap: Key Science Studies to Prepare for Soil Moisture Missions
6.0 Technology Roadmap: Key Technology Issues for a 10 km Mission
Soil moisture is the key state variable in hydrology: it is the switch that controls the proportion of rainfall that percolates, runs off, or evaporates from the land. It is the life-giving substance for vegetation. Soil moisture integrates precipitation and evaporation over periods of days to weeks and introduces a significant element of memory in the atmosphere/land system. There is strong climatological and modeling evidence that the fast recycling of water through evapotranspiration and precipitation is the primary factor in the persistence of dry or wet anomalies over large continental regions during summer. As a result, soil moisture is the most significant boundary condition that controls summer precipitation over the central U.S. and other large mid-latitude continental regions, and essential initial information for seasonal predictions.
Precise in situ measurements of soil moisture are sparse and each value is only representative of a small area. Remote sensing, if achievable with sufficient accuracy and reliability, would provide truly meaningful wide-area soil wetness or soil moisture data for hydrological studies over large continental regions.
Developing an effective soil moisture remote sensing system based on microwave radiometry requires the deployment of large antennas (or realization of a correspondingly large synthetic aperture) in order to achieve meaningful spatial resolution at the low microwave frequencies necessary to penetrate moderately dense vegetation. The primary objective of an experimental soil moisture measurement mission is to produce a synoptic global 3 to 5 year data set that will advance science.
An L band soil moisture measurement mission at a 30-50 km spatial resolution is desired as soon as possible. This mission would focus on Hydroclimatology. It will bring the necessary initial experience with the space deployment of a large low-frequency microwave radiometer. More importantly however, it will make possible those science applications that clearly demonstrate the significance of monitoring this variable. The ESA SMOS mission could satisfy most of the requirements of a Hydroclimatology mission and should be supported by NASA. A follow-on mission should augment the initial mission with the goal of a 10 km spatial resolution and increased temporal coverage. This will make possible the monitoring of soil moisture at finer scales compatible with the needs of Hydrometeorology. The soil moisture mission has enormous potential. It is comparable to TRMM and TOPEX at their origins.
The strategic importance of world water resources and food production make soil moisture a crucial variable for policy decisions. Since no regional or global soil moisture data sets currently exist, this mission will provide completely new information. The impact factor across hydrology, ecology, and the atmospheric and solid earth sciences would be enormous.
Given that soil moisture measurements would become available at regional to continental scales, the following specific science questions can be addressed by a soil moisture mission:
A spaceborne L band radiometer mission will be of significant value in sea surface salinity studies because L band is also the primary frequency needed for this application. In cold land processes L band data would contribute to the retrieval of information on the accumulation and depletion of snow over ice sheets. For surface water monitoring, the large difference in emissivity between land and water could provide information on flooding in large river basins. The high revisit interval may make the data very useful when merged with data from other sensors.
1. The Science Imperative
Soil moisture is the state variable of land surface hydrology. Although it constitutes a small percentage of global water, it is on a par with atmospheric water vapor in terms of its influence on the Earth's physical capacity to sustain life. Soil moisture is the fastest component of the water cycle (residence time of a few days) and plays a key role in partitioning precipitation and radiation at the land surface. Through its dominant influence on key physical processes, soil moisture is a variable that has always been required in many disciplinary and cross-cutting scientific and operational applications (e.g. ecology, biogeochemical cycles, climate monitoring, flood forecasting, etc.). To date there are neither in situ nor remote sensing systems that can provide direct estimates of global soil moisture fields. The only option has been to estimate this key hydrologic variable as a residual of water balance calculations.
Recent technological breakthroughs now allow the deployment of spaceborne sensors that can produce a data stream on soil moisture that is unprecedented. These measurements represent a quantum leap in NASA Earth observing capabilities because they fill a major gap in the data provided by the suite of sensors currently deployed to monitor the Earth. The scientific disciplines will have to fundamentally rethink the theoretical foundations of their models and procedures because for the first time observations of one of the main hydrological state variables will be available. Thus a soil moisture mission will represent an initiative for NASA's Earth Systems Enterprise (ESE) that has a tremendous science return for the investment. A soil moisture mission is, in this respect, a strategic mission.
Cross-Cutting Science Requirements for Soil Moisture Data
A NASA ESE soil moisture monitoring initiative will directly address the national priority of the U.S. Global Change Research Program (USGCRP) to develop improved capability to understand and predict the Earth's environment, especially for climate-sensitive sectors at regional scale. A new data stream on soil moisture will also substantially impact international science programs such as Global Energy and Water Cycle Experiment (GEWEX) and the Global Ocean-Atmosphere-Land System (GOALS) component of the Climate Variability and Predictability Program (CLIVAR) that are focused on the "fast" and "slow" components of climate variability. Recent reviews of these programs have consistently identified observing and characterizing soil moisture as a priority observing and science objective.
The National Research Council (NRC) (1995) review by the Board of Sustainable Development recommends that NASA use advanced technology on small satellites to supply soil moisture measurements from space. Similarly the NRC (1998a) review of the GEWEX Continental-Scale International Project (GCIP) has identified monitoring soil moisture as one of its six principal recommendations. GOALS has also identified measurement of soil moisture as a requirement for understanding and predicting climate variability (NRC, 1998b).
The lack of a global soil moisture observing system is one of the most glaring and pressing deficiencies in satellite remote sensing and climate research. The addition of observations of soil moisture to the current suite of NASA and international space missions will result in Earth science data products that are unprecedented. Ongoing programs such as GEWEX-GCIP and GOALS will be substantially impacted by soil moisture remote sensing.
Required Observational Scales
In all earth sciences natural variability is evident across a wide range of scales. For example, physical processes ranging in scale from molecular diffusion to planetary waves are responsible for the flux of heat, mass, and momentum. Different physical processes contribute to the total variability in a distinct range of scales. Therefore science questions and applications associated with each physical process need to know variations in relevant parameters at a given scale and higher.
Climate variations cause large-scale patterns in soil moisture and they are in turn influenced by soil moisture at the same scales (>30 km). In order to develop useful data sets to understand and predict hydroclimatic processes (such as large-scale droughts and changes in continental water cycling), variations in soil moisture fields resolved down to this scale are required. Other physical processes such as severe hydrometeorological events that cause flash floods are influenced by surface moisture conditions at the scale of convective storms and small basins (<10 km). Useful soil moisture data sets for understanding and predicting hydrometeorological events should resolve variations that occur at this scale.
In summary, whether it is hydroclimatic or hydrometeorological understanding and prediction, useful soil moisture data sets are those that resolve variations in surface moisture fields at scales that impact the phenomenon under consideration. There certainly are other processes such as soil texture variations and micro-topography that affect variability in soil moisture fields at scales down to a few meters and even the soil pore scale in some instances. Clearly remote sensing is not the ideal tool for the study of these physical processes even though they are important specialties in the hydrologic sciences.
A few examples of current science questions and applications that would be substantially impacted now by the availability of the proposed soil moisture data products are presented below.
How This Mission Addresses the ESE Science Questions
Is the climate changing in ways we can understand and predict?
This ESE priority issue addresses the broad scientific agenda of the USGCRP at the national level and the World Climate Research Program (WCRP) and International Geosphere-Biosphere Programme (IGBP) at the international level.
The redistribution of solar energy over the globe is central to the climatic system and its capacity to sustain life. Near-surface soil moisture serves a fundamental role in this redistribution through the energy associated with evaporation. There have been a number of numerical climate modeling studies that have demonstrated the sensitivity of climate to soil moisture (Walker and Rowntree, 1977; Rowntree and Bolton, 1983; Rind, 1982; Shukla and Mintz, 1982; Delworth and Manabe, 1988, 1989, 1993; and Mintz and Walker, 1993). The practical implication of this sensitivity is that the predictability of the climate system, whether it is for seasonal prediction or global change assessment, is dependent on robust characterization of soil moisture in the models. Access to global soil moisture estimates derived from spaceborne sensors at scales resolved by these numerical models of weather and climate (ranging from 30 to hundred of kilometers) will provide a powerful means to validate and improve the land component of these models.
The potential impact of surface soil moisture observations in GCMs is related to providing increased accuracy. Koster et al. (1999) provides an illustration of the enhanced predictability resulting from the correct knowledge of soil moisture fields. They found that with the added foreknowledge of land surface moisture conditions, precipitation predictability is enhanced in many continental regions, even in mid-latitudes. The land mainly contributes to predictability in the transition zones between dry and humid regions which coincide with areas where large-scale variations in soil moisture are strongest and a spaceborne soil moisture mission can provide valuable data.
The land surface boundary condition plays a key role in determining the diurnal evolution of the atmospheric boundary layer together with geographic and synoptic controls. In turn, the boundary layer largely controls the evolution of convection and the diurnal cycle of precipitation (again influenced by regional dynamic features). The key boundary layer parameters are the equivalent potential temperature and the lifting condensation level. The seasonal cycle of soil temperature is closely linked to the seasonal cycle of equivalent potential temperature (Betts and Ball, 1995), but superimposed on this is a large impact of soil moisture changes. High values of soil moisture lead, in the mean, to a larger evaporative fraction, a higher afternoon equivalent potential temperature, and a lower cloud-base. These favor the development of precipitating convection, which has a positive feedback on soil moisture. Large-scale studies using both climate and forecast models show this positive feedback, which can extend periods of drought or excess precipitation until the cycle is broken by changes in the planetary-scale wave pattern.
Can we understand and predict how terrestrial and marine ecosystems are changing?
This ESE initiative addresses scientific questions relevant to and in some cases critical for sustainable land management and ecosystem functioning. Plant function and the land components of global cycles of carbon, methane and nitrogen are linked with soil moisture at the surface and in the root zone. Transpiration by vegetation and soil microbial respiration rates are known to be dependent on water availability in the soil. Ecological models as well as global biogeochemical balances, as modeled or characterized by in situ and remote sensing of air chemistry, require estimates of soil moisture for closure. Data are required on a global scale and at resolutions that resolve biome differences (30 km and higher).
Can we improve our understanding of the processes and dynamics of the earth's surface and interior, and use this knowledge to assess and mitigate natural hazards?
This ESE initiative concerns the use of NASA technology to understand natural hazards and mitigate their consequences. A number of environmental disasters are so linked to soil moisture that better measurements would help assess the potential for damage from such events. These include floods, flash floods, infectious diseases, insect infestations, and landslides. Floods are, by a large margin, the most costly of natural hazards in terms of both human lives and property damages (NRC, 1996).
For example, the predictability of a recent severe flooding event in the midwestern United States has been shown to be highly sensitive to the characterization and tracking of soil moisture conditions across the region. Betts et al. (1996), Chen et al. (1996), Beljaars et al. (1996), and Paegle et al. (1996) consistently show that the hindcast skills of operational high-resolution Numerical Weather Prediction (NWP) and regional atmospheric model forecasts of the 1993 Upper Midwest U.S. flooding event (a disaster that produced $15 billion in damages) are improved with realistic soil moisture initial conditions. Wetter than normal soils preceding July 1993 caused greater recycling of local precipitation, leading to enhanced flooding.
Furthermore, Mitchell et al. (1996) and Mitchell (1999) demonstrate that the increase in skill associated with including estimates of soil moisture in NWP used in the U.S. is equivalent to the increase in skill when model resolution was doubled. Current NWP models have a 30 km resolution and in the future the enhanced models may achieve 10 km resolution. Soil moisture observations with the proposed satellite systems are scheduled in phase with this transition in NWP models and can provide most valuable initialization fields.
Can NASA assist in the development, implementation, testing and evaluation of new applications-oriented sensors that will help use the perspective and quantitative measurement capability of space-based observations for the public good?
Operational flood forecasting is currently based on limited measurements of rainfall and river stage, and in some cases on some type of soil moisture description usually in the form of an antecedent precipitation index. In many cases, poor forecasts are attributed to lack of information about the initial conditions, i. e., about soil moisture.
Flash flood and flood forecasting are other examples of how soil moisture remote sensing can directly support operational activities related to hazards mitigation. The U.S. National Weather Service (NWS) River Forecast Centers (RFCs) currently produce flash-flood guidance data products that are used by NWS local offices to issue flash-flood warnings and watches (NRC, 1996). These guidance products are essentially antecedent surface moisture fields at about 30 km scale and higher. Observed precipitation fields (principally from radar) are imposed on these fields in order to rapidly produce warnings and watches for the public. Availability of observed soil moisture at these same scales will substantially impact the operational practices of validating and calibrating flash-flood guidance products.
2.0 Historical Context for a Soil Moisture Mission
Passive microwave sensing (radiometry) has shown the greatest potential among remote sensing methods for satisfying the soil moisture measurement objectives of regional-to-global scale weather and climate forecasting. Measurements at 1 to 3 GHz are directly sensitive to changes in surface soil moisture, are little affected by clouds, and can penetrate moderate amounts of vegetation. They can also sense moisture in the surface layer to depths of 2 to 5 cm (depending on wavelength and soil wetness). With radiometry, the effect of soil moisture on the measured signal dominates over that of surface roughness (whereas the converse is true for radar).
The soil moisture mission is the logical consequence of a coordinated research program on retrieving soil moisture from microwave brightness (Figure 1). The approach evolved from small-scale studies using tower-mounted radiometers in the late 1960's and early 1970's (Edgerton et al., 1970), and truck- and aircraft-mounted radiometers during the late 1970's and early 1980's (Wang et al., 1983 and Jackson et al., 1984). With the development of the L band Push-Broom Microwave Radiometer (PBMR), it became the standard technique for soil moisture retrieval in the comprehensive airborne field campaigns of 1986 - 1992 (Jackson and O'Neill, 1987, Schmugge et al., 1992).
Real aperture radiometers like PBMR produced highly useful soil moisture products, but implementing that technology for large-aperture spaceborne systems was considered to be a significant challenge. Synthetic aperture technology was explored as a potential solution. An aircraft prototype for Earth-viewing applications was developed called the Electronically Scanned Thinned Array Radiometer (ESTAR) in the 1990's which proved to be equivalent in performance to traditional radiometers for sensing soil moisture (Le Vine et al., 1994, Jackson et al., 1993, Jackson et al., 1995). ESTAR was used in the Southern Great Plains (SGP97) hydrology experiment to map soil moisture accurately over an 11,000 km2 area nearly every day for a month (Figure 2) (Jackson et al., 1999).
The results from recent decades of modeling and experimentation have demonstrated that L band radiometry is the best technique for measuring soil moisture. They have also shown that a spaceborne mission is scientifically justified and that the hydrological community stands ready to use such data (1993 Moran Report, 1994 Tiburon Hydrology Workshop, 1998 Columbia Hydrology Workshop).
Only one low-frequency (< 3 GHz) spaceborne radiometer, the Skylab 1.4-GHz radiometer (S-194 mission), has ever flown in space. This radiometer operated on a short-duration mission in 1973 and 1974, providing nadir-only measurements at a spatial resolution of ~115 km. At nadir, horizontal and vertical polarization measurements should be the same. As part of this mission, Eagleman and Lin (1976) collected actual ground observations of soil moisture. This adds substantial credibility to the results. These observations and water balance simulations were used to study relationships between brightness temperature and soil moisture. The data were collected over the U.S. Great Plains during the summer. Their results are shown in Figure 3 and convincingly demonstrate the following key points: observed brightness temperature has a large dynamic range due to changing surface wetness conditions (55 K), there is a strong relationship between brightness temperature and soil moisture even at a resolution of 115 km, and the basic relationship is consistent with theory (model curve).
Since then, other higher frequency Earth-imaging microwave radiometers have operated in space, including the Scanning Multichannel Microwave Radiometer (lowest frequency 6.6 GHz) launched on the Seasat (1978) and Nimbus-7 (1978-87) satellites, and the Special Sensor Microwave Imager (lowest frequency 19.35 GHz) launched on the DMSP satellite series. Attempts have been made to utilize these sensors in soil moisture studies with some limited success (Choudhury and Golus, 1988, Jackson, 1997, Kerr and Njoku, 1990, Owe et al., 1992, and Wang, 1985). The Advanced Microwave Scanning Radiometer (AMSR) (lowest frequency 6.9 GHz) will be flown on the ADEOS-II (2000) and EOS-PM (2000) satellites. The capabilities of these higher frequency instruments are limited to soil moisture measurements over predominantly bare soil and in a very shallow surface layer (<1 cm). Data from these sensors cannot meet the requirements of the hydrology, weather, and climate communities in regions that include moderate vegetation cover.
The last sentence in the previous paragraph is particularly pertinent to the SSM/I instrument. Much effort has been devoted to extracting land information from this sensor, since it is an operational instrument and a series of these instruments have been in orbit since 1987. The SSM/I has shown only limited capacity to observe soil moisture from space (Lakshmi et al., 1997 and Jackson 1997). At its lowest frequency of 19.35 GHz the SSM/I is highly sensitive to even small amounts of vegetation which obscures the underlying soil. Large variations in soil moisture (e.g., flood/no-flood) in sparsely vegetated regions, and qualitative river flooding indices, are all that have been shown feasible using the SSM/I as reported in the peer-reviewed literature. For this reason, no serious plans for assimilating SSM/I-derived soil moisture exist within the operational forecasting and climate communities.
2.1 Underlying Physics
Among the microwave frequencies that are practical for radiometry, decades of experiment and theory have demonstrated that the frequencies near 1.4 GHz (L band) that lie within the protected radio astronomy band for hydrogen emission are best for sensing soil moisture (Jackson, 1993 and Schmugge et al., 1986). Figure 4 is a schematic of the components that comprise the microwave scene brightness at L band frequencies, and Figure 5 shows the corresponding brightness contributions of sky, canopy, and soil for an incidence angle of 0o. The sky brightness originates from the cosmic background (at L band it is less than 2 K) and atmospheric contributions from clouds and water vapor are considered negligible.
The L band single scattering albedo for vegetation canopies is generally less than 0.02 (Kerr and Wigneron, 1994; Ulaby et al., 1983; O'Neill et al., 1996) so that these canopies can be modeled as an absorbing layer. The optical thickness of crop or grassland canopies is typically much less than 1 and can be estimated from knowledge of the equivalent water column in the canopy. This requires ancillary data from an independent source such as a satellite-derived vegetation index (Jackson, 1993). Alternatively, dual-polarized (V and H) measurements at incidence angles of 40° or greater or other passive and active channels may be used to estimate the vegetation optical thickness (Njoku and Li, 1999 and Njoku et al., 1999).
2.2 Heritage of Large Antennas
The long electromagnetic wavelength needed for soil moisture remote sensing, and the perceived need for a spatial resolution less than 100 km, will require antennas with a large aperture size. Work that was principally funded by space structures research began in the mid 1970s to address this problem. As this progressed, it became clear that mechanical scanning was required in order to provide global coverage if such apertures were to have value for remote sensing of soil moisture. It further became clear that mechanical scanning of such a large aperture posed feasibility problems using the technology available at that time. In order to circumvent these problems the idea of a pushbroom radiometer was advanced in the late 1970s. This concept places many identical microwave receivers in the focal plane of the reflector antenna, such that each receiver observes at a specific pixel cross-track to the orbital motion of the large reflector antenna. An image is thus developed through the forward motion of the spacecraft. The swath width can be increased by adding identical receivers. This can add up to 50 or more for a modest footprint size across a wide swath. The pushbroom concept was demonstrated with flight units developed by NASA Langley Research Center and the Technical University of Denmark.
The difficulty with the pushbroom concept was the continued growth in size that was required in order to accommodate large number of receivers. A 10 km resolution with a 1000 km swath required an antenna aperture with dimensions of 25 by 50 m. As a result of the size growth issue, aperture synthesis was pursued as an alternative in the early 1980s. Although a large antenna with 25 m by 25 m dimensions was still required, a one dimensional thinning offered a means to reduce the weight and volume of the antenna in order to better accommodate a space launch. The reduction in antenna area dimensions strongly suggested that the antenna could be easily packaged for launch, and deployed after the orbit was achieved. The concept also incorporated phased array technology which allowed for electronic rather than mechanical scanning. Since this work began, it appears that a road map can be drawn to eventually place a two dimensional synthetic aperture antenna into space for further reduction in mass and volume.
Recently, classified work in the development and space flight of large filled-aperture antennas has been released on low-mass deployable technologies, indicating the feasibility of using such antennas for microwave remote sensing applications. Thus, the choice of antenna concepts has come full circle. Rotating and light weight filled apertures with multiple receivers placed near the focus of the reflector are again being explored as an option. Because of these developments, parallel technology investigations are being pursued in aperture synthesis (HYDROSTAR, SMOS) and in filled aperture systems (OSIRIS).
2.3 Large Footprint Scale and Heterogeneity Issues
There are two issues that have to be considered when dealing with coarse resolution data such as that generated by passive microwave sensors, namely scaling and heterogeneity. In scaling it is necessary to establish that microwave/soil moisture relationships developed over small spatial areas using truck and aircraft radiometers over the past 20 years can be scaled up to the larger areas and coarser resolutions more typical of space missions. The recent SGP97 experiment was conducted in part to address this issue. L band truck radiometer data (SLMR) and measured soil moisture taken over small footprints in two representative test sites were compared with ESTAR brightness temperature and measured soil moisture averaged over the surrounding 75 km2 area. Despite the obvious variations in surface conditions and non-homogeneity across this area, both ESTAR and SLMR responded the same way to changes in soil moisture, indicating that the microwave-soil moisture relationship is equally valid at point and small regional scales. A similar conclusion was reached by Kim and England (1998) in comparing higher microwave frequency data from SSM/I and a tower-mounted radiometer in the Arctic tundra. In addition, ESTAR data from three different experiments over the Little Washita watershed (a 630 km2 area in the southern part of the SGP test region) demonstrate that the microwave-soil moisture relationship is repeatable and predictable at the mission resolution (Jackson et al., 1999).
Using this same data set, Drusch et al. (1999) showed that upscaling was a valid assumption for L band brightness temperatures over this region for footprints ranging from 0.8 km to 31.2 km. This means that the average derived from individual smaller footprints is the same as that of an individual larger footprint. Basic physics verified at smaller scales can be applied to satellite scale footprints.
Heterogeneity involves the issue of whether large footprint observations over heterogeneous terrain will represent an area-wide average corresponding to the sensor's footprint. The relationship between this "sensor-averaged" or "effective" value and the simple weighted average of the higher resolution within-pixel components has been the subject of research. In investigating the mixed pixel problem from a modeling standpoint, Njoku et al. (1996) determined that at L band the difference between the sensor average and the component average was very small - less than 0.2 kg/m2 for plant water content and ~1-2% for mean soil moisture. Liou et al. (1998) also concluded that mixed vegetation scales linearly at L band - it is the total amount of biomass and not its spatial distribution within the pixel that is important in affecting the microwave response. Similarly, Galantowicz et al. (1999) showed that soil texture variations also scale linearly at L band - the RMS error in soil moisture retrieved from TB in the sensor footprint compared to that retrieved from the individual homogeneous soil regions within the footprint was less than 1%. These papers all point to the conclusion that subpixel heterogeneity in soil texture, vegetation, and land cover will not significantly impact the retrieval of soil moisture information from satellite brightness temperature data.
2.4 Linking Surface Moisture to the Profile Soil Moisture
A microwave remote sensing system provides a direct measurement of the surface soil moisture. Most hydrologic investigations require complete profile soil moisture. Physically based relationships and models can be used with the surface observations in estimating the profile soil moisture (Jackson, 1980). The science and operational apparatus to perform this task has been in development for a number of years now. Approaches of data assimilation have been used to solve both the soil moisture profile inference and downscaling problems based on merging multispectral remote sensing, in situ observations, and models of the soil-vegetation-atmosphere continuum. Results of Houser et al. (1998) and Entekhabi et al. (1994) among others indicate that the necessary procedures are in place to derive higher-level data products from the L band radiometry data stream. Programs such as the Land Data Assimilation System http://ldas.gsfc.nasa.gov could ultimately utilize these data and provide the surface-profile linkage.
3.0 Mission Rationale
A satellite soil moisture mission is built on four key arguments:
As described in Section 1, there is a wide
range of science questions with varying requirements that can be addressed
by a soil moisture mission. Some of these requirements can be met by the
proposed SMOS mission, while others will need additional technology investment
leading to the next generation sensors. Therefore, the acquisition of global
soil moisture data sets should be viewed as a sequence of two missions
(see Table 1), with SMOS addressing a specific set of science questions
and the second mission being more technically challenging and enabling
an extended range of science applications.
|Table 1. Soil Moisture Mission Parameters|
|Launch Readiness Date||2005 (SMOS)||2008|
|Spatial Resolution (km)||50||10|
|Revisit Interval (days)||3||2-3|
|Accuracy (% Soil Moisture)||4||4|
|Technology Alternatives||SMOS||Aperture synthesis, mesh antenna, unfocused radar as part of active / passive approach|
The Hydroclimate mission can be achieved using techniques that are currently feasible. At the present time, the European Space Agency (ESA) has selected the Soil Moisture Ocean Salinity (SMOS) mission for an extended Phase A study. This mission can satisfy the demands of the Hydroclimate mission. With a limited investment NASA could partner with ESA and support a U.S. SMOS science team and/or make a more substantial contribution to the design and success of the SMOS mission.
SMOS (Table 2) is a dual-polarized 2-D
aperture synthesis radiometer providing ~50 km resolution. It has sufficient
swath at the nominal orbit altitude to insure complete global coverage
with no data gaps. Current projections indicate that it would be ready
for launch in 2005.
|Table 2. Summary of the SMOS Hydroclimate Soil Moisture Mission|
||Antenna Technology||2-D synthetic aperture|
|Frequencies (GHz)||1.4 passive|
|Imaging Technique||Multi-angular 50-55 deg.|
|Spatial Resolution (km)||50|
|Revisit Interval (d)||2-3|
|Algorithm Status||Ground validation|
|Technology Issues||Receivers, antenna coupling|
|Ongoing Programs||Aircraft/ground prototypes|
SMOS will be launched into a sun synchronous (6 a.m. ascending), circular, 757 km orbit. Raw measuring performances then are: 30 to more than 90 km for ground resolution, 0.8 to 2 K for radiometric sensitivity, 2 to 3 days for temporal sampling, depending upon latitude, nature of the target and location within the instrument field of view. The mission minimum duration is 3 years (5 years expected).
SMOS aims at providing, over the open ocean, global salinity maps with an accuracy better than 0.1 PSU every few days, with a 200 km spatial resolution; over the land surfaces, global maps of soil moisture, with an accuracy better than 0.035 m3/m3 every 3 days, with a space resolution better than 60 km, as well as vegetation water content with an accuracy of 0.2 kg/m2. Under the ESA mission, in order to process SMOS data for soil moisture, corrections due to atmospheric, ionospheric and galactic effects have to be applied. Over land surfaces, knowledge of the temperature is needed. Both the Jackson (1993) and Njoku and Li (1999) algorithms could be applied.
Implementing a soil moisture mission that could satisfy the demands of hydrologic processes and hazards studies in a Hydrometeorology mission is a much more challenging problem because the needed spatial resolutions are on the order of 10 km or better. In addition to all of the science and algorithm development needed, this mission would require substantial advances in technology. Although technically challenging, with appropriate technology investment it is anticipated that such a mission would be launch-ready in ~2007-2008. The science and application feasibility of this mission could be explored through studies in conjunction with earlier missions such as the Hydroclimate mission (discussed above) and the NASA proposed LightSAR mission.
It is possible that a SAR system could contribute to the advanced higher resolution soil moisture mission as a complement to the primary passive instrument system. SAR systems have been flown in space since 1978. While these satellites have proven the maturity of SAR technology, these systems typically have much higher resolutions, and consequently data rates, than what would be required for a mission with a resolution on the order of 1 - 10 km. In addition, the typical repeat cycles are on the order of 30 days or more, making the data of limited value for hydrologic process studies, and the current soil moisture retrieval algorithms using SAR data are less accurate than desired in handling vegetation effects.
Several technologies are under development to reduce the cost of SAR missions. These include the miniaturization of flight electronics and the development of inflatable antennas. Using inflatable antennas would significantly reduce the launch mass and volume of such a SAR mission, reducing the overall cost. Processing global data sets of SAR data for conventional SAR systems is a daunting task, made even more challenging if the processing has to be repeated every few days. While computer processing speeds have increased dramatically over the past decade, it still is not realistic to think that conventional SAR processing could handle global data sets on a regular basis in the next 5-10 years. The soil moisture SAR system, however, does not require full conventional SAR processing to achieve the required resolution. Utilizing unfocussed SAR processing, one can easily achieve resolutions on the order of 1 - 10 km with many (on the order of 50 or more) looks. Unfocussed SAR processing is several orders of magnitude less demanding computationally than conventional SAR processing.
5.0 Science Roadmap: Key Science Studies to Prepare for Soil Moisture Missions
Although much research has been conducted to date, as we begin to focus on specific missions it appears that there are still some outstanding questions that need to be answered on basic physical relationships and algorithms. These evaluations should be conducted under over a range of ground conditions. Field experiments are needed which include conditions that have not yet been thoroughly evaluated.
A key component of a global soil moisture mission is a series of validation sites where high quality ground data will be collected consistently. The role of existing soil moisture measurement and temperature networks (i.e. Oklahoma, Illinois, Russia, USDA NRCS and Ameriflux) should be explored in the context of their ability to monitor the mean moisture content within an area of about two satellite resolution cells. The number of sites needed for validation also needs to be determined. If current instrumented networks are inadequate, it is critical that action be taken immediately to establish additional sites. Additional supporting research on scaling point to areal measurements will be required to assess the potential bias in the mean statistics provided by the networks. It is critical that this work be initiated prior to the launch of the AMSR instruments. Not only would such action benefit the AMSR product validation but it would also allow the testing and comparison of alternative designs prior to a dedicated global soil moisture mission.
Provide Airborne Simulators of Mission Instruments and Aircraft Support
Airborne simulators for each proposed space instrument need to be built for pre-mission studies and mission-current validation flights. The pre-mission studies will be used to validate the retrieval algorithm and provide examples of mission-like data. During the mission the airborne simulator will be used for validation of the spaceborne sensor and for conducting algorithm and application experiments. At this time one of the most significant shortcomings is the lack of a two dimensional synthetic aperture instrument that is capable of collecting data in exactly the same manner as potential satellites. At a minimum, support for the current satellite prototypes (ESTAR and the JPL PALS) should be continued while a 2-D simulator is pursued. NASA aircraft are critical to this aspect of a soil moisture mission.
Simulations Studies to Define Application Requirements and Benefits
A shortcoming in the past has been the lack of a focused program to fully integrate the potential soil moisture information with the needs of different user applications. The most obvious application areas for use of remotely sensed soil moisture are agriculture, water resources and basin hydrology, flood forecasting, and mesoscale weather prediction. Although it is intuitively obvious that measured soil moisture should be able to provide added value in these areas, this needs to be demonstrated quantitatively with the following simulation studies:
Error Performance Simulations
As described in the text, there are strong science justifications for both a 30-50 km/2-3 day and a 10 km/2-3 day soil moisture mission. Some technology issues apply to both missions while others are specific to the advancements necessary to enable the more challenging 10 km mission.
Achieving Global Daily Coverage
All soil moisture mission proposals to date are based on global 3 day coverage at a resolution of 30-50 km. A long term goal desirable in a Hydrometeorology mission is to obtain daily coverage of changes in surface soil moisture. Trade studies of alternative strategies and required technologies needed to achieve this goal should be identified.
Achieving 10 km spatial resolution from a 600 km orbit requires an antenna aperture (real or synthesized) of about 25 m. Achieving 2-3 day global coverage will require multiple orbiting sensors, or a single sensor with wide swath (~1500 km or greater). Wide swath can be achieved by moving to a higher orbit, but at the cost of reduced spatial resolution for the same antenna size.
Frequency Tradeoffs for Spatial Resolution
The feasibility of using an S and C band frequency combination for soil moisture retrieval should be considered as a possible path to a 10 km 1-2 day coverage soil moisture mission. This would significantly reduce the required antenna size and thus minimize some of the engineering risks and costs of a 10 km mission. The effects of this frequency combination instead of an L band channel on sensitivity to soil moisture, RFI interference, and other factors would need to be studied. Some of these issues could be pursued using C-band AMSR data and the Hydroclimatology mission.
The next step in the application of this technology for remote sensing of soil moisture at high spatial resolutions of 10 km or better is to develop aperture synthesis in two dimensions as a logical follow-on to our experience in successfully mapping soil moisture with the 1-D approach. Specific tasks include:
Mesh antennas of diameter 12.5 m, using two different designs-- "articulated-rib" and "perimeter-truss", will be launched on commercial geostationary communication satellites by the end of year 2000. Industry studies have shown that antennas of similar designs are feasible at 25-m, and possibly larger, although these studies did not address the technical problems to be encountered when spinning such large structures in orbit, which would be necessary to produce a conical scanning mapping radiometer.
The challenge in going to larger (25 m) antenna diameters for a rotating system is in the control and stability of the system. Higher spatial resolution on the ground requires faster spin rates for contiguous mapping. Multiple beams, feeds and radiometers can be employed to keep the rotation rate low.
The use of multiple frequencies and combined radiometry-radar will broaden the applications of lightweight real-aperture antennas for hydrological applications, including soil moisture, snow, and frozen-ground monitoring at higher resolution. Multichannel passive and active integrated feeds and electronics design studies, at frequencies up to Ku-band, should be pursued for these applications.
Specific studies to be performed include:
Beljaars, A. C. M., P. Viterbo, M. J. Miller, and A. J. Betts (1996) The anomalous rainfall over the United States during July 1993: Sensitivity to land surface parameterization and soil moisture anomalies. Mon. Wea. Rev., 124:362-383.
Betts, A., J. Ball, and A. Beljaars (1993) Comparison between the land surface response of the ECMWF model and the FIFE-1987 data. Q.J.R. Meteorol. Soc., 119:975-1001.
Betts, A., F. Chen, K. Mitchell, and Z. Janjic (1996) Assessment of land-surface and boundary-layer models in 2 operational versions of the Eta model using FIFE data. Mon. Weather Review, 124:1480-1498.
Chen, F., Z. Janjic, and K. Mitchell (1996) Impact of atmospheric surface layer parameterization in the new land-surface scheme of the NCEP mesoscale Eta numerical model. Boundary-Layer Meteorology.
Choudhury, B. J. and R. E. Golus (1988) Estimating soil wetness using satellite data. Int. J. Rem. Sens., 9:251-1257.
DeFries, R.S., M. Hansen, M., J. R. G. Townshend, R. and Sohlberg (1998) Global land cover classifications at 8km spatial resolution: the use of training data derived from Landsat imagery in decision tree classifiers. International Journal of Remote Sensing, 19:13141-3168.
Delworth, T. and S. Manabe (1988) The influence of potential evaporation on the variabilities of simulated soil wetness and climate. J. Climate, 1:523-547.
Delworth, T. and S. Manabe (1989) The influence of soil wetness on near-surface atmospheric variability. J. Climate, 2:1447-1462.
Delworth, T. and S. Manabe (1993) Climate variability and land-surface processes. Advances in Water Resources, 16:3-20.
Drusch, M., E. F. Wood, and C. Simmer (1999) Up-scaling effects in passive microwave remote sensing: ESTAR 1.4 GHz measurements during SGP'97. Geophysical Research Letters, 26:879-882.
Eagleman, J. R. and W. C. Lin (1976) Remote sensing of soil moisture by a 21-cm passive radiometer. J. of Geophysical Research, 81:3660-3666.
Edgerton, A., D. Trexler, G. Poe, A. Stogryn, S. Sakamoto, J. Jenkins, D. Meeks, and F. Soltis (1970) Passive microwave measurements of snow, soils, and oceanographic phenomena. Tech. Rep. No. 6, SD 9016-16, Aerojet-General Corp., El Monte, CA.
Entekhabi, D., H. Nakamura, and E. G. Njoku (1994) Solving the inverse problem for soil moisture and temperature profiles by sequential assimilation of multifrequency remotely sensed observations. IEEE Trans. Geosci. Rem. Sens., 32:438-448.
Galantowicz, J., D. Entekhabi, and E. Njoku (1999) Estimation of soil type heterogeneity effects in the retrieval of soil moisture from radiobrightness. IEEE Trans. on Geosci. and Remote Sensing, in press.
Houser, P. R., W. J. Shuttleworth, J. S. Famiglietti, H. V. Gupta, K. Syed, and D. C. Goodrich, (1998) Integration of soil moisture remote sensing and hydrologic modeling using data assimilation, Water Resources Research, 34:3405-3420.
Jackson, T. J. (1980) Profile soil moisture from surface layer measurements. Journal of the Irrigation and Drainage Division of the ASCE, 106:81-92.
Jackson, T. J. (1993) Measuring surface soil moisture using passive microwave remote sensing. Hydrological Processes, 7:139-152.
Jackson, T. J. (1997) Soil moisture estimation using SSM/I satellite data over a grassland region. Water Resources Research, 33:1475-1484.
Jackson, T.J. and P. E. O'Neill (1987) Temporal observations of surface soil moisture using a passive microwave sensor. Remote Sensing of Environment, 21: 281-296.
Jackson, T. J. and T. J. Schmugge (1991) Vegetation effects on the microwave emission from soils. Remote Sensing of Environ., 36:203-212.
Jackson, T. J., T. J. Schmugge, and P. E. O'Neill (1984a) Passive microwave remote sensing of soil moisture from an aircraft platform. Remote Sensing of Environment, 14:135-152.
Jackson, T. J., D. M. Le Vine, A Griffis, D. C. Goodrich, T. J. Schmugge, C. T. Swift, and P. E. O'Neill (1993) Soil moisture and rainfall estimation over a semiarid environment with the ESTAR microwave radiometer. IEEE Trans. on Geoscience and Remote Sensing, 31:836-841.
Jackson, T. J., D. M. Le Vine, C. T. Swift, and T. J. Schmugge (1995) Large scale mapping of soil moisture using the ESTAR passive microwave radiometer. Remote Sensing of Environment, 53:27-37.
Jackson, T. J., D. M. Le Vine, A. Y. Hsu, A. Oldak, P. J. Starks, C. T. Swift, J. Isham and M. Haken (1999) Soil moisture mapping at regional scales using microwave radiometry: the Southern Great Plains hydrology experiment. IEEE Trans. on Geoscience and Remote Sensing, in press.
Kerr, Y. H. and E. G. Njoku (1990) A semiempirical model for interpreting microwave emission from semiarid land surfaces as seen from space. IEEE Trans. Geosci. Rem. Sens., 28:384- 393.
Kerr, Y. H. and J. P. Wigneron (1994) Vegetation models and observations - review, In: Passive Microwave Remote Sensing of Land Atmosphere Interactions, ESA/NASA International Workshop - 1993 (Saint-Lary), p.317-344, B. Choudhury, Y. Kerr, E. Njoku, P. Pampaloni (Eds), VSP, Utrecht.
Kim, E. and A. England (1998) Land surface process modeling and passive microwave remote sensing of Arctic tundra. Proc. of IGARSS '98, Seattle, WA, July, 1998, pp. 1849-1851.
Koster, R. and M. Suarez (1996) The influence of land surface moisture retention on precipitation statistics. Journal of Climate, 9:2551-2567.
Koster, R. D., M. J. Suarez, and M. Heiser (1999) Variance and predictability of precipitation at seasonal-to-interannual time scales, submitted to J. Hydrometeorology.
Lakshmi, V., E.F.Wood and B.J.Choudhury (1997) Evaluation of SSM/I satellite data for regional soil moisture estimation over the Red River Basin. Journal of Applied Meteorology, Vol. 36:1309-1328
Le Vine, D. M., A. Griffis, C. T. Swift, and T. J. Jackson (1994) ESTAR: a synthetic microwave radiometer for remote sensing applications. Proceedings of the IEEE, 82:1787-1801.
Liou, Y.A., E.J. Kim, and A.W. England (1998) Radiobrightness of prairie soil and grassland during drydown simulations," Radio Science, 33:259-265.
Mitchell, K. (1999) Personal Communication.
Mitchell, K., T. Black, E. Rogers, C. Peters, Z. Janjic, Y. Xue, C. Ropelewski, J. Horel, A. Vernekar, F. Baer, C. Nobre, and E. Berberry (1996) GCIP improvements to Eta model benefit studies world wide. GEWEX News, vol. 6, no. 2, WCRP.
Mintz, Y. and G. Walker (1993) Global fields of soil moisture and land surface evapotranspiration derived from observed precipitation and surface air temperature. Journal of Applied Meteorology, 32:1305-1334.
National Research Council (1995) A review of the USGCRP and NASA's MTPE/EO. National Academy Press, Washington., DC.
National Research Council (1996) Towards a new National Weather Service: assessment of hydrologic and hydrometeorological operations and services. National Academy Press, Washington, D.C.
National Research Council (1998a) Global energy and water cycle experiment (GEWEX) continental-scale international project (GCIP): a review of progress and opportunities. National Academy Press, Washington, D.C.
National Research Council (1998b) A scientific strategy for U.S. participation in the global ocean-atmosphere-land system (GOALS) component of the climate variability and predictability programme (CLIVAR). National Academy Press, Washington, D.C.
Njoku, E. G. and L. Li (1999) Retrieval of Land Surface Parameters Using Passive Microwave Measurements at 6 to 18 GHz. IEEE Trans. Geosc. Rem. Sens., 37:79-93.
Njoku, E., S. Hook, and A. Chehbouni (1996) Effects of surface heterogeneity on thermal remote sensing of land parameters. in: Scaling-Up In Hydrology Using Remote Sensing (J. B. Stewart, E. T. Engman, R. A. Feddes, and Y. Kerr, Eds.), Wiley, Chichester.
Njoku, E. G., Y. Rahmat-Samii, J. Sercel, W. J. Wilson, and M. Moghaddam (1999) Evaluation of an inflatable antenna concept for microwave remote sensing of soil moisture and ocean salinity. IEEE Trans. On Geoscience and Remote Sensing, 37:63-78.
O'Neill, P., N. Chauhan, and T. Jackson (1996) Use of active and passive microwave remote sensing for soil moisture estimation through corn. Int. Journal of Remote Sensing, 17:1851-1865.
O'Neill, P., A. Hsu, T. Jackson, and C. Swift (1998) Ground-based microwave radiometer measurements during the Southern Great Plains '97 experiment. Proc. of IGARSS '98, Seattle, WA, July, 1998, pp. 1843-1845.
Owe, M., A. A. van de Griend, and A. T. C. Chang (1992) Surface moisture and satellite microwave observations in semiarid southern Africa. Water Resources Res., 28:829-839.
Paegle, J., K. Mo, and J. Nogues-Paegle (1996) Dependence of simulated precipitation on surface evaporation during the 1993 US summer floods. Mon. Weather Review, 124:345-361.
Rind, D. (1982) The influence of ground moisture conditions in North America on summer climate as modeled in the GISS GCM. Mon. Weather Rev., Vol. 110:1487-1494.
Rowntree, P. and J. Bolton (1983) Simulation of the atmospheric response to soil moisture anomalies over Europe. Q.J.R. Meteorol. Soc., 109:501-526.
Schmugge, T. J., P. E. O'Neill, and J. R. Wang (1986) Passive microwave soil moisture research. IEEE Trans. on Geoscience and Remote Sensing, GE-24:12-22.
Schmugge, T. J., T. J. Jackson, W. P. Kustas, and J. R. Wang (1992) Passive microwave remote sensing of soil moisture: results from HAPEX, FIFE and MONSOON 90. ISPRS Journal of Photogrammetry and Remote Sensing, 47:127-143.
Shukla, J. and Y. Mintz (1982) Influence of land-surface evapotranspiration on the Earth's climate. Science, 215:1498-1500.
Ulaby, F. T., R. K. Moore, and A. K. Fung (1983) Microwave remote sensing: active and passive, Vol. I. Artech House, Dedham, MA.
Walker, J., and P. Rowntree, (1977) The effect of soil moisture on circulation and rainfall in a tropical model. Q.J.R. Meteorol. Soc., 103:29-46.
Wang, J. R. (1985) Effect of vegetation on soil moisture sensing observed from orbiting microwave radiometers. Rem. Sens. Environ., 17:141-151.
Wang, J. R., P. E. O'Neill, T. J. Jackson,
and E. T. Engman (1983) Multifrequency measurement of thermal microwave
emission from soils: The effects of soil texture and surface roughness.
IEEE Trans. on Geoscience and Remote Sensing, GE-21:44-55.
Return to Top