Importance of Soil Moisture and Retrieval Processes
Soil moisture (SM) is an important variable in hydrometeorological models. It affects numerous phenomena and gets affected at the same time, like precipitation, runoff, vegetation growth, infiltration to groundwater, and evapotranspiration. Also, soil moisture is highly related to agriculture, one of the most important industries in human society.
There have been various methods to understand soil moisture of a specific place at a specific time, and there are three methods in big criteria. Satellite-based methods are the most prevailing method recently, and also a really powerful method. It can cover a large area at one time, varying by orbit type but it can cover the entire Earth as well. However, satellite-based retrieval has a limitation of resolution, for both temporal and spatial resolution. Because satellites orbit the Earth, it is impossible to get continuous data for a specific area. It is true that stationary orbit can compensate for this drawback, then there is no reason to use a satellite instead of direct investigation. For major soil moisture missions, like Soil Moisture Active Passive (SMAP) from NASA and Soil Moisture and Ocean Salinity (SMOS) have 2 to 3 days of revisiting time, and sometimes it may be longer due to unexpected occasions making data unavailable.
Also, compared to other methods, spatial resolution is still not as good as the others. Because the satellite orbits the Earth, it has a long distance from the investigating place and the sensor attached in the satellite. First, the distance is long, and second, there is an atmosphere between them. Because of these two factors, there is always an issue for uncertainty in satellite-based datasets. Because the signal is traveling long distances, there is a limit of each mission’s spatial resolution. The resolution varies by missions, for instance SMAP and SMOS missions use a 25 km EASEgrid system. But still, there have been attempts to overcome this limitation, some datasets offer better resolution, for instance SMAP L3 Enhanced dataset offers a 9 km EASEgrid system.
Second method is model analysis. There have been a lot of hydrometeorological models depicting the Earth, and lots of them are having good performance. However, the biggest problem for the model is that it cannot be used only, since the model data cannot predict completely accurately for the unexpected events, especially when having more and more extreme events due to climate change. Thus, inconsistent accuracy of the model is a big limitation.
Final method is ground observation, which is directly investigating the place. It is pretty evident that this method has a coverage issue. Of course the Earth is so large, and because of lots of realistic reasons, it is impossible to cover all of them with sensors and retrieve soil moisture. Not only the soil moisture retrieval, all the ground observation has the same problem. Nonetheless, it has the best accuracy as well as consistent and continuous data, so it is a powerful method if investigating on a small area.
Core Validation Site (CVS)
Because of uncertainty in satellite-based dataset, lots of sites have been built for validating the retrieved dataset from satellites. By comparing the data from satellite and ground observation, it is possible to calibrate the system. Unfortunately, no CVS site has been built in Korea, so advanced level of soil moisture retrieval in Korea is restricted a lot. There are lots of sites with sensors, but not sophisticated as NASA’s CVS guideline has mentioned.
According to NASA’s requirements for automated soil measurements, there should be 12 sensors in a 33 km grid cell, 8 sensors in a 9 km grid cell, 6 sensors in a 3 km grid cell and 1 km grid cell. Currently, 9 km, 3 km, and 1 km grid cell candidate sites are settled, so the soil moisture sensor will be deployed there to retrieve soil moisture directly.
Below figure, Fig. a shows the total research area in this project. The CVS is mainly located in Hakgyo-myeon, Hampyeong-gun, Jeollanam-do. Some part of the research area includes Hampyeong-eup, Daedong-myeon, Eomda-myeon in Hampyeong-gun and Munpyeong-myeon, Dasi-myeon, Gongsan-myeon, Donggang-myeon in Naju-si.
Fig. b shows the 9 km grid cell, and the candidate sites are depicted as red. Fig. d shows the 3 km grid cell, and the candidate sites are depicted as green. Also, the green cell in Fig. b is the same cell in Fig. d. Finally, Fig. c shows the 1 km grid cell, and the candidate sites are depicted as white. Samely, the white cell in Fig. d is the same cell in Fig. c.
When the CVS is fully prepared, all these locations will have sensors to retrieve soil moisture in different places. The location for each sensor is selected when having homogeneous vegetation, so that errors and uncertainties occurring from different types of vegetation and soil can be minimized. Also the selection of the sites are advised by NASA researchers, who actually run the CVS sites in other countries. For further information, look up (Link) for NASA researchers visiting GIST.
Radiometer and other Instruments
In the CVS site, there will be one radiometer center, and one drone-based radiometer, and several sensors in different places. Designed by the European Space Agency (ESA), ESA L-Band Radiometer (ELBARA-III) will be the main radiometer here. L-Band (around 1.41 GHz microwave signal) is used in a lot of soil moisture satellite missions. Not only ELBARA-III, there will be two more radiometers, which is the Portable L-Band Radiometer (PoLRa). One of them will be attached in ELBARA-III tower, and the other will be attached to a drone, enabling it to retrieve airborne data. For further information about radiometers, look up for (link).
In each grid cell, TEROS 12-6 soil moisture sensors will be deployed, retrieving soil moisture, soil temperature, electrical conductivity, pressure and logger temperature. Among them, soil moisture and soil temperature will be important values for CVS dataset, working as a validation data there. Also, the sensor is connected to a controller which runs in a solar panel, expecting to run without disturbed by manpower.
Goal of the project
Basic goal of the project is to build a CVS site, which will be the first CVS site in Korea. With all the instruments prepared, it is expected to retrieve brightness temperature(TB) from radiometer, and soil moisture directly from the TEROS sensor. Next, a model converting from TB value to soil moisture will be built, and lots of ancillary data, such as soil temperature, soil information(clay fraction / sand fraction / silt fraction), soil roughness, soil porosity and emissivity will also be input data of the model. With this model, it is expected to retrieve soil moisture value from the radiometer.
