Satellite remote sensing has brought an unprecedented opportunity to hydrologic community. The availability of satellite data helps us understand some of the main controls on the runoff production in any region in the world. Yes! That is the power of satellite remote sensing.
In this new study, we have investigated the role of antecedent soil moisture, event-scale rainfall and vegetation on the runoff generation. Our study domain is the state of Iowa where we have used SMAP satellite data, MODIS vegetation index and radar rainfall to find the main factors that control the runoff production.
Main finding is that although there are issues with the sensitivity of L-band microwave to vegetation, still we have found a strong relationship soil moisture estimated from SMAP (Soil Moisture Active Passive) satellite data. Interestingly, we have shown that we should expect higher runoff ratios for the rainfall events with larger total rainfall depth. We combine soil moisture and rainfall into one measure so called "Soil Moisture Deficit Normalized Rainfall". We defined it as the ratio of event-scale rainfall to available space in the soil prior to intiation of the rainfall.
Well, lets say you have a bucket and you want to fill it with water. if you fill it more than the available space, then it will overflow. On the other hand, if you fill it less than the available space, then you will be able to carry it easier. Similarly, if rainfall is more than available space in the top soils, then the water will runoff. "Deficit-normalized rainfall" can help us find the total runoff. However, we save this for another day!
We finally study the annual cycles of runoff ratio and vegetation by using 17 years of data from MODIS (Moderate Resolution Imaging Spectroradiometer). We found that vegetation plays a major role in decreaseing the runoff ratio. The findings of our study can be used for hydrologic assessment of watershed management over long time periods. Also, the impact of vegetation on runoff is significant that we need to think and incorporate these dynamics in the hydrologic modeling frameworks so as to account for the dynamics of the runoff.
We estimate the soil moisture variability for basin and SMAP sub-grid using radar rainfall and soil properties.
We drive the statistical information for dry-down and wetting of soils and we use this information to estimate the spatial and temporal variability of soil moisture at the basin and satellite sub-grid. Our methods and results is useful where higher resolution soil moisture data is needed. On the other hand, we show that the variance (and standard deviation) of a bounded variable should be bounded. However, our methodology leads to an estimation of variability and skewness of soil moisture.
As there is precipitation event over the basin, the soil moisture tends to increase and the mean value increases. At the same time, standard deviation of soil moisture decreases for wetter conditions. This phase is more rapid than dry-downs and changes the state of the soils faster. I have created these a video for a demonstration of how the soil moisture distribution changes over time as the wetting and dry-down events occur.
For example, in here I am showing an animation of the soil moisture distribution over time for Turkey River basin, located in North East of State of Iowa, USA.
While satellite-based soil moisture estimations are coarse in spatial resolution, our methods could be used for estimating variability of soil moisture at a given pixel or watershed.
For more details, you can download our paper in the attachment.