Several recent studies have highlighted the potential of Actively Heated Fiber Optics (AHFO) for high resolution soil moisture mapping. In AHFO, the soil moisture can be calculated from the cumulative temperature (T_cum), the maximum temperature (T_max), or the soil thermal conductivity determined from the cooling phase after heating (λ). This study investigates the performance of the T_cum, T_max and λ methods for different heating strategies, i.e., differences in the duration and input power of the applied heat pulse. The aim is to compare the three approaches and to determine which is best suited to field applications where the power supply is limited. Results show that increasing the input power of the heat pulses makes it easier to differentiate between dry and wet soil conditions, which leads to an improved accuracy. Results suggest that if the power supply is limited, the heating strength is insufficient for the λ method to yield accurate estimates. Generally, the T_cum and T_max methods have similar accuracy. If the input power is limited, increasing the heat pulse duration can improve the accuracy of the AHFO method for both of these techniques. In particular, extending the heating duration can significantly increase the sensitivity of T_cum to soil moisture. Hence, the T_cum method is recommended when the input power is limited. Finally, results also show that up to 50% of the cable temperature change during the heat pulse can be attributed to soil background temperature, i.e., soil temperature changed by the net solar radiation. A method is proposed to correct this background temperature change. Without correction, soil moisture information can be completely masked by the background temperature error.

This study addresses two critical barriers to the use of Passive Distributed Temperature Sensing (DTS) for large-scale, high-resolution monitoring of soil moisture. In recent research, a particle batch smoother (PBS) was developed to assimilate sequences of temperature data at two depths into Hydrus-1D to estimate soil moisture as well as soil thermal and hydraulic properties. However, this approach was limited to bare soil and assumed that the cable depths were perfectly known. In order for Passive DTS to be more broadly applicable as a soil hydrology research and remote sensing soil moisture product validation tool, it must be applicable in vegetated areas. To address this first limitation, the forward model (Hydrus-1D) was improved through the inclusion of a canopy energy balance scheme. Synthetic tests were used to demonstrate that without the canopy energy balance scheme, the PBS estimated soil moisture could be even worse than the open loop case (no assimilation). When the improved Hydrus-1D model was used as the forward model in the PBS, vegetation impacts on the soil heat and water transfer were well accounted for. This led to accurate and robust estimates of soil moisture and soil properties. The second limitation is that, cable depths can be highly uncertain in DTS installations. As Passive DTS uses the downward propagation of heat to extract moisture-related variations in thermal properties, accurate estimates of cable depths are essential. Here synthetic tests were used to demonstrate that observation depths can be jointly estimated with other model states and parameters. The state and parameter results were only slightly poorer than those obtained when the cable depths were perfectly known. Finally, in situ temperature data from four soil profiles with different, but known, soil textures were used to test the proposed approach. Results show good agreement between the observed and estimated soil moisture, hydraulic properties, thermal properties, and observation depths at all locations. The proposed method resulted in soil moisture estimates in the top 10 cm with RMSE values typically 3/m³. This demonstrates the potential of detecting the spatial variability of soil moisture and properties in vegetated areas from Passive DTS data.

This study investigates the potential of estimating the soil moisture profile and the soil thermal and hydraulic properties by assimilating soil temperature at shallow depths using a particle batch smoother (PBS) using synthetic tests. Soil hydraulic properties influence the redistribution of soil moisture within the soil profile. Soil moisture, in turn, influences the soil thermal properties and surface energy balance through evaporation, and hence the soil heat transfer. Synthetic experiments were used to test the hypothesis that assimilating soil temperature observations could lead to improved estimates of soil hydraulic properties. We also compared different data assimilation strategies to investigate the added value of jointly estimating soil thermal and hydraulic properties in soil moisture profile estimation. Results show that both soil thermal and hydraulic properties can be estimated using shallow soil temperatures. Jointly updating soil hydraulic properties and soil states yields robust and accurate soil moisture estimates. Further improvement is observed when soil thermal properties were also estimated together with the soil hydraulic properties and soil states. Finally, we show that the inclusion of a tuning factor to prevent rapid fluctuations of parameter estimation, yields improved soil moisture, temperature, and thermal and hydraulic properties.

This study demonstrated a new method for mapping high-resolution (spatial: 1 m, and temporal: 1 h) soil moisture by assimilating distributed temperature sensing (DTS) observed soil temperatures at intermediate scales. In order to provide robust soil moisture and property estimates, we first proposed an adaptive particle batch smoother algorithm (APBS). In the APBS, a tuning factor, which can avoid severe particle weight degeneration, is automatically determined by maximizing the reliability of the soil temperature estimates of each batch window. A multiple truth synthetic test was used to demonstrate the APBS can robustly estimate soil moisture and properties using observed soil temperatures at two shallow depths. The APBS algorithm was then applied to DTS data along a 71 m transect, yielding an hourly soil moisture map with meter resolution. Results show the APBS can draw the prior guessed soil hydraulic and thermal properties significantly closer to the field measured reference values. The improved soil properties in turn remove the soil moisture biases between the prior guessed and reference soil moisture, which was particularly noticeable at depth above 20 cm. This high-resolution soil moisture map demonstrates the potential of characterizing soil moisture temporal and spatial variability and reflects patterns consistent with previous studies conducted using intensive point scale soil moisture samples. The intermediate scale high spatial resolution soil moisture information derived from the DTS may facilitate remote sensing soil moisture product calibration and validation. In addition, the APBS algorithm proposed in this study would also be applicable to general hydrological data assimilation problems for robust model state and parameter estimation.

Surface heat fluxes interact with the overlying atmosphere and play a crucial role in meteorology, hydrology, and climate change studies, but in situ observations are costly and difficult. It has been demonstrated that surface heat fluxes can be estimated from assimilation of land surface temperature (LST). One approach is to estimate a neutral bulk heat transfer coefficient (CHN) to scale the sum of turbulent heat fluxes, and an evaporative fraction (EF) that represents the partitioning between fluxes. Here the newly developed particle batch smoother (PBS) is implemented. The PBS makes no assumptions about the prior distributions and is therefore well-suited for non-Gaussian processes. It is also particularly advantageous for parameter estimation by tracking the entire prior distribution of parameters using Monte Carlo sampling. To improve the flux estimation on wet or densely vegetated surfaces, a simple soil moisture scheme is introduced to further constrain EF, and soil moisture observations are assimilated simultaneously. This methodology is implemented with the FIFE 1987 and 1988 data sets. Validation against observed fluxes indicates that assimilating LST using the PBS significantly improves the flux estimates at both daily and half-hourly timescales. When soil moisture is assimilated, the estimated EFs become more accurate, particularly when the surface heat flux partitioning is energy-limited. The feasibility of extending the methodology to use remote sensing observations is tested by limiting the number of LST observations. Results show that flux estimates are greatly improved after assimilating soil moisture, particularly when LST observations are sparse.