3.4 Implications
Existing methods for direct field measurement of soil hydraulic properties remain complex, time-consuming, costly, and significant spatial and temporal variability challenges the possibility of extensive measurements (Ma et al., 2018). However, the spatial patten, as indicated by the EOF analysis, is the relatively stable. This stability attribute of the spatial pattern has implied that it is possible to continuously assess the soil moisture distribution in a catchment. In addition to the spatial coverage maps, adding in the long-term monitoring of both surface and subsurface soil moisture provides a comprehensive picture of the spatial-temporal pattern of soil moisture dynamics across the whole area and allow the identification of factors which influence it through time. A unique long-term real time soil moisture data set was previously used to identify local dominant hydrological processes and its time dynamics. In this perspective, our approach becomes more effective given that the long-term monitored site is characterized as a time-stable location via a time-stability analysis (Zhao et al., 2010). Our approach also has the capability to assimilate additional data sources, e.g., remote sensed data at this time-stable site. Given the high accuracy of the soil moisture monitoring, the time-resolution soil moisture patterns over an area could be obtained by selecting a temporally stable monitoring site, which is useful in ground truthing of a remotely sensed footprint for validation of simulation modelling results (Zhao et al., 2010). Given the importance of soil moisture in Earth’s land surface interactions and to a large range of applications, one can appreciate that its accurate estimation is critical in addressing key practical challenges such as food security, sustainable planning and management of water resources. The launch of new, more sophisticated satellites strengthens the development of innovative research approaches and scientific inventions that will result in ground-breaking advancements.
One goal of this study was to lay the foundation for the design of cost-effective real-time soil moisture monitoring networks that fill in the gap between point sensors and traditional manual measurements or even remote sensing values. Our study is representative of a novel approach with the potential benefits for an effective soil moisture monitoring network design within the study area, determining the spatiotemporal statistics of the observed soil moisture fields, and the use of a spatial regression procedure in data merging. It is more realistic to observe a difference between developed maps as surface conditions evolve. We believe that this combination reflects more adequately the basin heterogeneity and complex interactions between soil moisture and topographic attributes. Although there are the simple linear data transfer methods that have been applicable to this type of, our approach may accommodate different data analysis methods, such as a multi-step regression method based on the EOF analysis (Temimi et al., 2010). Once the tasks within our approach have been completed, the EOF-based transfer method may be used as a foundation any region and/or date under the assumptions that the identified empirical relationships will be valid for the application conditions.
The present soil-landscape has been shaped through a combination of long- and short-time processes, and this history can provide some clues to project future changes. However, linking long-term and slow processes with shorter-term and fast processes remains one challenge (Ma et al., 2018). While mapping depicts the spatial distribution of soil-landscape relationships, as indicated by the dominant EOF patterns; monitoring captures the temporal dynamics of pedologic and hydrologic properties, as indicated by the profiled data dynamics (Ma et al., 2018). Given the spatial-extensive data benefits of traditional mapping, and the temporal-extensive data benefits of traditional monitoring, the presented data-integrated method may provide a justifiable basis for the combination of mapped and monitored data, as well as a conceptual basis for the coupling of slow and fast processes. Firstly, bridging mapping with monitoring is very helpful in the dynamic mapping of hydropedologic functional units (Ma et al., 2018). Secondly, mapping provides information to aid in optimal site selection for monitoring (Zhao et al., 2010). Thirdly, mapping and monitoring supplies essential data for the calibration and validation of modeling, and may help provide additional information for a more holistic, refined and predictive management of soil and water resources (Guo et al., 2019). Our approach provides an essential set of tools to evaluate the improvement of data use. We assumed that the relationship between different data sources remains the same over time, but suggest that future studies verify this behavior.