# Soil Moisture in the Biebrza Wetlands Retrieved from Sentinel-1 Imagery

^{1}

^{2}

^{*}

## Abstract

**:**

## 1. Introduction

_{veg}, and of the underlying soil σ°

_{soil}. The WCM model was adopted by Dabrowska-Zielinska et al. [9] for agricultural fields. The separation of the soil and vegetation components is not straightforward due to the complex interactions between them, which simultaneously affect SAR backscatter. The signal strongly depends on the type of vegetation, the amount of moisture, and the type of ecosystem [9]. Wetlands are characterized by deep peat layers, and it is not possible to compare agriculture ecosystems to wetlands, which are wet and very different. Thus, the models derived for wetlands have to be treated separately from models that are designated for agriculture soils and agriculture vegetation.

^{3}m

^{−3}and 0.059 m

^{3}m

^{−3}for the first and second methods, respectively. El Hajj et al. [23] used a neural network technique to develop an operational method for soil moisture estimates in agricultural areas based on the synergistic use of Sentinel-1 and Sentinel-2 data. They found that VV polarization alone as well as both VV and VH provides better accuracy on the soil moisture calculation than VH alone. The method developed by them could be applied for agricultural plots with an NDVI lower than 0.75 and allows for the soil moisture estimates with an accuracy of approximately 5 vol. %. Baghdadi at al. [24] applied the Water Cloud Model for estimating surface soil moisture of crop fields and grasslands from Sentinel-1/2 data. They simulated the soil contribution (moisture content and surface roughness) applying Integral Equation Model and used NDVI values as the vegetation descriptor. They obtained that the soil contribution to the total radar signal is large in VV polarization when soil moisture is between 5 and 35 vol. %, and NDVI between 0 and 0.8. Tomer et al. [25] developed an algorithm to retrieve surface soil moisture based on the Cumulative Density Function Transformation of multi-temporal RADARSAT-2 backscattering coefficient. The algorithm, which was tested in a semi-arid tropical region in South India and validated with the in situ data showed RMSE of soil moisture estimates ranging from 0.02 to 0.06 m

^{3}m

^{−3}depending on soil information used and development of vegetation. Dabrowska-Zielinska et al. [26] conducted an investigation on soil moisture monitoring in the Biebrza Wetlands using Sentinel-1 data, and found, that LAI dominates the influence on σ° when soil moisture is low. They developed models for soil moisture assessment under different wetland vegetation habitat types (non-forest communities) applying VH polarization (R

^{2}= 0.70 to 0.76). There are not many studies for wetlands SM retrieval applying S-1 data, as can be seen from the literature review. Most of the publications refer to agriculture crops or bare soils. The difference and the ratio of the VH and VV backscatter as the proxy of vegetation conditions has been recently studied and published by several researchers. Vreugdenhil et al. [27] examined Sentinel-1 VV and VH backscatter and their ratio VH/VV to monitor crop conditions with special reference to vegetation water content (VWC) of agriculture crop. Greifeneder et al. [28] demonstrated that the ratio of VH/VV calculated from AQUARIUS L-band scatterometer allows a good compensation of vegetation dynamics for the retrieval of soil moisture. Hosseini et al. [29] used RADARSAT-2 to estimate Leaf Area Index (LAI) for corn and soybeans fields. They found high correlation coefficients between ground measured and estimated LAI values, when dual like-cross polarizations were used (either HH–HV or VV–HV). Also, it has been found that RADARSAT-2 (HH-HV) can be used for the retrieval of soil moisture and the total biomass, while RADARSAT-2 (VV-HV) can be used for the retrieval of the biomass of the wheat heads [30].

## 2. Materials and Methods

#### 2.1. Study Area

^{2}, the river length is 155 km, and its mean flow is 35.3 m

^{3}s

^{−1}. The Wetlands are flooded annually in the spring, and besides precipitation, flooding is the main supply of moisture into the peat soil. The weather in the Biebrza River Valley is one of the coolest in Poland—the mean year daily temperature is 6.5 °C. The mean sum of the yearly precipitation ranges between 550–650 mm, and is one of the lowest in Poland. The length of the growing season is less than 200 days, and this is one of the shortest in Poland. Generally, summer is warm but short; winter is cold and long. The coldest month is January, with a mean temperature of −4.2 °C, and with temperatures dropping as low as −50 °C. Snow cover can last up to 140 days. July is the warmest month in the Biebrza Valley, with mean temperatures of 17.5 °C, and with temperatures increasing up to 35.3 °C. The length of the summer ranges between 77–85 days [32].

#### 2.2. In Situ Data

^{−2}) were measured. These data supported the SM analysis with ancillary information about the variables influencing the SAR signal (biomass, vegetation conditions).

#### 2.3. Satellite Data

#### 2.4. Methods

^{®}, Norcross, GA, USA).

#### 2.4.1. Water Cloud Model with the Least Squares Method

_{veg}and of the underlying soil σ°

_{soil}[36]:

_{veg}+ τ

^{2}σ°

_{soil}

_{veg}= A V

_{1}cos(θ) (1 − τ

^{2})

^{2}= exp(−2B V

_{2}/cos(θ))

^{2}—two way attenuation through the canopy: V

_{1}and V

_{2}are descriptors of the canopy, A and B are fitted parameters of the model that depend on the vegetation descriptor and the radar configuration. As the vegetation descriptors (V

_{1}and V

_{2}), the NDVI values derived from MODIS data were taken. The B parameter is connected with the density of vegetation and its strength of the attenuation during the growing season. For the specific, homogeneous area, we can assume the fixed value of B and apply linearized nonlinear method to solve the WCM model (instead of nonlinear iterative methods). Figure 4 presents the simulation of the strength of attenuation depending on NDVI values for different values of B parameter.

_{soil}) is a linear function. It was assumed that in early spring at the wetlands area the soil has dominated impact on backscatter. Therefore we applied modified WCM, where σ°

_{soil}(Equation (1)) was represented by measured SM values. The measurements were conducted during two full years at even time interval, so the relation soil-vegetation can be assumed to be well represented. The following two components of data were designed to describe the effect of the vegetation and the underlying soil on σ° VH value: τ

^{2}* SM, and (1 − τ

^{2})* cos (θ)* NDVI. The first component represents the interaction of the incident radiation between the vegetation and the underlying soil. τ

^{2}reduces the impact of the soil on backscatter when the vegetation cover is dense. τ

^{2}takes the value from 0–1 and is inversely proportional to the vegetation index and to the incidence angle. The second component describes the remaining part of the backscatter that depends on the vegetation canopy covering the soil. The parameters of the model with σ° VH as a dependent variable, and τ

^{2}* SM and (1 − τ

^{2})* cos (θ)* NDVI as independent variables, were estimated by applying the Least Squares Method. Data were limited to the vegetation season, i.e., from 60–300 days of each year. The form of modified WCM model is the following:

^{2}SM + c(1 − τ

^{2}) cos(θ) NDVI

#### 2.4.2. Vegetation Descriptors

_{NIR}− R

_{RED})/(R

_{NIR}+ R

_{RED}),

_{RED}—spectral reflectance in the red spectrum, R

_{NIR}—spectral reflectance in the near-infrared spectrum. For calculating NDVI all pixels with the spectral reflectance values larger than 0 and lower than 10,000 (16 bit unsigned integer) were taken. Then, from Band 3 (Surface Reflectance 250 m State flags) of MOD09Q1 product the pixels flagged as: water, clouds/cloud shadows, and snow/ice were extracted and applied to NDVI images. The values of spectral reflectance were the ratios of the reflected radiation over the incoming radiation in each spectral channel individually (albedo); hence, the NDVI takes on values between 0–1.

#### 2.4.3. Statistical Analyses

^{2}(coefficient of determination), MAPE (Mean Absolute Percentage Error), MPE (Mean Percentage Error), RMSE (Root Mean Square Error), and MBE (Mean Bias Error). The data were checked for the normal distribution and significance prior to all analyses. Validation of the retrieved SM values against the in situ measurements was preformed based on the RMSE error.

## 3. Results

#### 3.1. Correlation between σ° Calculated from S-1 and Soil Moisture Measured at Different Depths

#### 3.2. Impact of Vegetation on σ° Calculated from S-1 under Different Soil Moisture Conditions

#### 3.3. Impact of Soil Moisture on σ° Calculated from S-1 under a Quasi-Constant NDVI

#### 3.4. Compatibility of Seasonal Trends in the Course of the Vegetation Descriptor NDVI, and the σ° Difference VH−VV and Ratio VV/VH

- 1
- Using the NDVI as a vegetation descriptor
- 2
- Substituting the NDVI by the index σ° VH−VV and the index σ° VV/VH

#### 3.5. Soil Moisture Retrieval Using σ° from Sentinel-1 and NDVI from MODIS

^{2}= exp(−NDVI/cos(θ))

^{2}SM + 14.7(1 − τ

^{2}) cos(θ) NDVI

^{2}= 0.85; p < 0.0000; N = 147; Std. Err. = 0.79 dB, for ascending orbit.

^{2}= 1 and σ°

_{veg}= 0, the Equation (7) takes the following form: σ° VH = −28.3 + 0.2* SM. For the early spring measurements, when the vegetation has not started yet to grow, estimated equation has the following form: σ° VH = −34.4 + 0.21*SM, where R = 0.89; N = 34. In both simulated and estimated equations, the regression slope that means sensitivity, is the same. The intercept parameters which are connected with roughness of soil and vegetation cover, differ. This is the measure of the difference between the soil, theoretically bare, according to model (Equation (7)) and our assumption.

^{2}SM + 12.3(1 − τ

^{2}) cos(θ) NDVI

^{2}= 0.82; p < 0.0000; N = 170; Std. Err. = 0.84 dB, for ascending orbit.

^{2})* cos(θ)* NDVI)/(0.2* τ

^{2})

#### 3.6. Soil Moisture Retrieval Using the σ° Indices from Sentinel-1

^{2}= exp(−2(σ° VV/VH)/cos (θ))

^{2}*SM and (1 − τ

^{2})*cos (θ)*σ°(VH–VV)

^{2}. Then, σ° VH was modeled according to Model 2 applying linearized nonlinear regression method.

^{2}SM − 0.14(1 − τ

^{2}) cos(θ) σ°(VH–VV)

^{2}

^{2}= 0.82; p < 0.000; N = 252; Std. Err. = 0.70 dB, (Figure 12), for ascending orbits.

^{2}= 0.002.

^{2}= 1; constant a = −18.9 is the state of balance between the impact of vegetation and the underlying soil on σ° VH (SM about 50 vol. %, Figure 13). Under σ° VV < 0 the attenuation factor τ

^{2}(Equation (10)) is always less than 1, so the sensitivity does not reaches the value of 0.33, it is lower. Theoretically, sensitivity of SAR backscatter to soil moisture increases when the ratio σ° VV/VH decreases. Figure 14 shows the periods under low vegetation conditions.

^{2}for each track separately, the range of sensitivity of σ° VH backscatter was calculated. For the satellite track 29 (θ = 43°10′), the highest sensitivity was 0.084 dB/vol. % and the lowest was 0.029 dB/vol. %, while for the satellite track 131 ((θ = 35°13′) − 0.096 dB/vol. % and 0.036 dB/vol. %, respectively. It is compatible with the results when the NDVI from optical data were used (Figure 10 and Figure 11). For low SM there is the increase of σ° VH. For high values of SM, there is the attenuation of the beam by vegetation. Model 2 can be applied in all weather conditions, independently of sky conditions, on which the acquisition of optical images depends.

^{2}) cos(θ) σ°(VH−VV)

^{2}/(0.33 τ

^{2})

## 4. Discussion

## 5. Conclusions

## Author Contributions

## Funding

## Conflicts of Interest

## References

- Available online: https://www.biebrza.org.pl/lang,2 (accessed on 6 December 2018).
- Turbiak, J.; Miatkowski, Z. Assessment of organic mass mineralization rate in deeply drained peat-muck soil based on losses of soil mass and CO
_{2}emission. ITP Water Environ. Rural Areas**2016**, 16, 73–85. [Google Scholar] - Huang, W.; Hall, S.J. Elevated moisture stimulates carbon loss from mineral soils by releasing protected organic matter. Nat. Commun.
**2017**, 8, 1774. [Google Scholar] [CrossRef] [PubMed][Green Version] - Dembek, W.; Oswit, J.; Rycharski, M. Torfowiska i torfy w Pradolinie Biebrzy. In Przyroda Biebrzanskiego Parku Narodowego; Dyrcz, A., Werpachowski, C., Eds.; Biebrzanski Park Narodowy: Osowiec-Twierdza, Poland, 2005; pp. 33–58. [Google Scholar]
- Kornelsen, K.C.; Coulibaly, P. Advances in soil moisture retrieval from synthetic aperture radar and hydrological applications. J. Hydrol.
**2013**, 476, 460–489. [Google Scholar] [CrossRef] - Allen, C.T.; Ulaby, F.T. Modelling the Polarization Dependence of the Attenuation in Vegetation Canopies. In Proceedings of the Geoscience and Remote Sensing Symposium (IGARSS’84), Strasbourg, France, 27–30 August 1984; pp. 119–124. [Google Scholar]
- Le Toan, T.; Lopes, A.; Huet, M. On the relationships between radar backscattering Coefficient and Vegetation Canopy Characteristics. In Proceedings of the Geoscience and Remote Sensing Symposium (IGARSS’84), Strasbourg, France, 27–30 August 1984; pp. 155–160. [Google Scholar]
- Attema, E.P.; Ulaby, F.T. Vegetation modeled as a water cloud. Radio Sci.
**1978**, 13, 357–364. [Google Scholar] [CrossRef] - Dabrowska–Zielinska, K.; Inoue, Y.; Kowalik, W.; Gruszczynska, M. Inferring the effect of plant and soil variables on C- and L-band SAR backscatter over agricultural fields, based on model analysis. Adv. Space Res.
**2007**, 39, 139–148. [Google Scholar] [CrossRef] - Mattia, F.; Satalino, G.; Dente, L.; Pasquariello, G. Using a priori information to improve soil moisture retrieval from ENVISAT ASAR AP data in semiarid regions. IEEE Trans. Geosci. Remote Sens.
**2006**, 44, 900–912. [Google Scholar] [CrossRef] - Balenzano, A.; Satalino, G.; Pauvels, V.; Mattia, F. Soil moisture retrieval from dense temporal series of C-band SAR data over agricultural sites. In Proceedings of the 2011 IEEE International Geoscience and Remote Sensing Symposium (IGARSS), Vancouver, BC, Canada, 24–29 July 2011. [Google Scholar] [CrossRef]
- Baghdadi, N.; El Hajj, M.; Zribi, M.; Fayad, I. Coupling SAR C-band and optical data for soil moisture and leaf area index retrieval over irrigated grasslands. IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens.
**2015**, 9, 1–15. [Google Scholar] [CrossRef] - Baghdadi, N.; Choker, M.; Zribi, M.; El Hajj, M.; Paloscia, S.; Verhoest, N.; Lievens, H.; Baup, F.; Mattia, F. A new empirical model for Radar scattering from bare surfaces. Remote Sens.
**2016**, 8, 920. [Google Scholar] [CrossRef] - Choker, M.; Baghdadi, N.; Zribi, M.; El Hajj, M.; Paloscia, S.; Verhoest, N.; Lievens, H.; Mattia, F. Evaluation of the Oh, Dubois and IEM models using large dataset of SAR signal and experimental soil measurements. Water
**2017**, 9, 38. [Google Scholar] [CrossRef] - Pasolli, L.; Notarnicola, C.; Bertoldi, G.; Bruzzone, L.; Remelgado, R.; Greifeneder, F.; Niedrist, G.; Della Chiesa, S.; Tappeiner, U.; Zebisch, M. Estimation of soil moisture in mountain areas using SVR technique applied to multiscale active radar images at c-band. IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens.
**2015**, 8, 262–283. [Google Scholar] [CrossRef] - Mattia, F.; Satalino, G.; Balenzano, A.; Pauwels, V.; DeLathauwer, E. GMES Sentinel-1 Soil Moisture Retrieval Algorithm Development; European Space Agency contract n. 4000101352/10/NL/MP/ef; Final Report; European Space Agency: Paris, France, 2011. [Google Scholar]
- Balenzano, A.; Mattia, F.; Satalino, G.; Pauvels, V.; Snoeij, P. SMOSAR algorithm for soil moisture retrieval using Sentinel-1 data. In Proceedings of the IEEE International Geoscience and Remote Sensing Symposium (IGARSS), Munich, Germany, 22–27 July 2012. [Google Scholar] [CrossRef]
- Paloscia, S.; Pettinanto, S.; Santi, E.; Notarnicola, C.; Pasolli, L.; Reppucci, A. Soil moisture mapping using Sentinel-1 images: Algorithm and preliminary validation. Remote Sens. Environ.
**2013**, 134, 234–248. [Google Scholar] [CrossRef] - Santi, E.; Paloscia, S.; Pettinato, S.; Fontanelli, G. Application of artificial neural networks for the soil moisture retrieval from active and passive microwave spaceborne sensors. Int. J. Appl. Earth Obs. Geoinf.
**2016**, 48, 61–73. [Google Scholar] [CrossRef] - Kasischke, E.; Bourgeau-Chavez, L.; Rober, A.; Wyatt, K.; Waddington, J.; Turetsky, M. Effects of soil moisture and water depth on ERS SAR backscatter measurements from an Alaskan wetland complex. Remote Sens. Environ.
**2009**, 113, 1869–1873. [Google Scholar] [CrossRef] - Lang, M.W.; Kasischke, E.S.; Prince, S.D.; Pittman, K.W. Assessment of C-band synthetic aperture radar data for mapping and monitoring Coastal Plain forested wetlands in the Mid-Atlantic Region, U.S.A. Remote Sens. Environ.
**2008**, 112, 4120–4130. [Google Scholar] [CrossRef] - Gao, Q.; Zribi, M.; Escorihuela, M.J.; Baghdadi, N. Synergetic Use of Sentinel-1 and Sentinel-2 Data for Soil Moisture Mapping at 100 m Resolution. Sensors
**2017**, 17, 1966. [Google Scholar] [CrossRef] [PubMed] - El Hajj, M.; Baghdadi, N.; Zribi, M.; Bazzi, H. Synergic Use of Sentinel-1 and Sentinel-2 Images for Operational Soil Moisture Mapping at High Spatial Resolution over Agricultural Areas. Remote Sens.
**2017**, 9, 1292. [Google Scholar] [CrossRef] - Baghdadi, N.; El Hajj, M.; Zribi, M.; Bousbih, S. Calibration of the Water Cloud Model at C-Band for Winter Crop Fields and Grasslands. Remote Sens.
**2017**, 9, 969. [Google Scholar] [CrossRef] - Tomer, S.K.; Al Bitar, A.; Sekhar, M.; Zribi, M.; Bandyopadhyay, S.; Sreelash, K.; Sharma, A.K.; Corgne, S.; Kerr, Y. Retrieval and Multi-scale Validation of Soil Moisture from Multi-temporal SAR Data in a Semi-Arid Tropical Region. Remote Sens.
**2015**, 7, 8128–8153. [Google Scholar] [CrossRef][Green Version] - Dabrowska-Zielinska, K.; Budzynska, M.; Tomaszewska, M.; Malinska, A.; Gatkowska, M.; Bartold, M.; Malek, I. Assessment of Carbon Flux and Soil Moisture in Wetlands Applying Sentinel-1 Data. Remote Sens.
**2016**, 8, 756. [Google Scholar] [CrossRef] - Vreugdenhil, M.; Wagner, W.; Bauer-Marschallinger, B.; Pfeil, I.; Teubner, I.; Rüdiger, C.; Strauss, P. Sensitivity of Sentinel-1 Backscatter to Vegetation Dynamics: An Austrian Case Study. Remote Sens.
**2018**, 10, 1396. [Google Scholar] [CrossRef] - Greifeneder, F.; Notarnicola, C.; Hahn, S.; Vreugdenhil, M.; Reimer, C.; Santi, E.; Paloscia, S.; Wagner, W. The Added Value of the VH/VV Polarization-Ratio for Global Soil Moisture Estimations from Scatterometer Data. IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens.
**2018**, 11, 3668–3679. [Google Scholar] [CrossRef] - Hosseini, M.; McNairn, H.; Merzouki, A.; Pacheco, A. Estimation of Leaf Area Index (LAI) in corn and soybeans using multi-polarization C- and L-band radar data. Remote Sens. Environ.
**2015**, 170, 77–89. [Google Scholar] [CrossRef] - Hosseini, M.; McNairn, H. Using multi-polarization C- and L-band synthetic aperture radar to estimate biomass and soil moisture of wheat fields. Int. J. Appl. Earth Obs. Geoinf.
**2017**, 58, 50–64. [Google Scholar] [CrossRef] - Available online: http://www.biebrza.org.pl/index.php (accessed on 6 December 2018).
- Silakowski, M. Charakterystyka fizyczno-geograficzna-Klimat. In Biebrzanski Park Narodowy; Sienko, A., Grygoruk, A., Eds.; Biebrzanski Park Narodowy: Osowiec-Twierdza, Poland, 2003; pp. 17–22. [Google Scholar]
- Musial, J.P.; Dabrowska-Zielinska, K.; Kiryla, W.; Oleszczuk, R.; Gnatowski, T.; Jaszczynski, J. Derivation and validation of the high resolution satellite soil moisture products: A case study of the Biebrza Sentinel-1 validation sites. Geoinf. Issues
**2016**, 8, 37–53. [Google Scholar] - Dorigo, W.; Wagner, W.; Hohensinn, R.; Hahn, S.; Paulik, C.; Xaver, A.; Gruber, A.; Drusch, M.; Mecklenburg, S.; Oevelen, P.V.; et al. The International Soil Moisture Network: A data hosting facility for global in situ soil moisture measurements. Hydrol. Earth Syst. Sci.
**2011**, 15, 1675–1698. [Google Scholar] [CrossRef] - Vermote, E. MOD09Q1 MODIS/Terra Surface Reflectance 8-Day L3 Global 250m SIN Grid V006 [Data set]. NASA EOSDIS LP DAAC.
**2015**. [Google Scholar] [CrossRef] - Prevot, L.; Champion, I.; Guyot, G. Estimating surface soil moisture and leaf area index of a wheat canopy using a dual—Frequency (C and X bands) scatterometer. Remote Sens. Environ.
**1993**, 46, 331–339. [Google Scholar] [CrossRef] - Rao, S.; Sahadevan, D.; Wadodkar, M.; Nagaraju, M.; Chattaraj, S.; Joseph, W.; Rajankar, P.; Sengupta, T.; Venugopalan, M.; Das, S.; et al. Soil Moisture Model with Multi Angle and Multi Polarisation Risat-1 Data. ISPRS Ann. Photogramm. Remote Sens. Spat. Inf. Sci.
**2014**, 2, 145. [Google Scholar] [CrossRef] - Narvekar, P.S.; Entekhabi, D.; Seung-Bum, K.; Njoku, E.G. Soil Moisture retrieval using L-band radar observations. IEEE J. Trans. Geosc. Rem. Sens.
**2015**, 53, 6. [Google Scholar] [CrossRef] - Available online: http://iopscience.iop.org/article/10.1088/1755-1315/105/1/012083/pdf (accessed on 6 December 2018).

**Figure 1.**Location of S-1 soil moisture sites at the Biebrza Welands overlapped to the Geoportal maps image (www.geoportal.gov.pl).

**Figure 5.**Relationship between the NDVI and σ° VH for the SM values measured at a 5 cm depth > 60 vol. % at the grassland site.

**Figure 6.**Relationship between the NDVI and σ° VH for the SM values measured at a 5 cm depth < 30 vol. % at the grassland site.

**Figure 7.**Temporal evolution of the NDVI and σ° VH−VV during the vegetation season of 2016 on the grassland site.

**Figure 10.**Impact of NDVI on σ° VH under various levels of soil moisture (SM) according to Model 1a.

**Figure 11.**Impact of NDVI on σ° VV under various levels of soil moisture (SM) according to Model 1b.

**Figure 13.**Impact of vegetation described by σ° VV/VH on σ° VH for different SM values according to Model 2.

**Figure 15.**Comparison between the soil moisture retrieved by the inversion of Model 2 according to Equation (12) (IGiK (Institute of Geodesy and Cartography) product) and soil moisture measured at a 5 cm depth (sm) by the Decagons GS3 sensors at the grassland site.

**Figure 16.**Comparison between the soil moisture retrieved by the inversion of Model 2 according to Equation (12) (IGiK (Institute of Geodesy and Cartography) product), and soil moisture measured at a 5 cm depth (sm) by the Decagons GS3 sensors at the marshland site.

**Table 1.**Local incidence angles for selected S-1 orbit passes (A-ascending, D-descending) and tracks.

Pass/Track | Marshland Incidence Angle | Grassland Incidence Angle |
---|---|---|

A/29 | 43.49° | 43.10° |

A/131 | 35.59° | 35.13° |

D/80 | - | 45.65° |

D/153 | 38.57° | 38.18° |

**Table 2.**Pearson’s correlation (R values) for the marshland site between σ° VH and VV from S-1 and soil moisture (GS3), measured in situ at three depths: 5, 10, and 20 cm.

Marshland 2015–2017 Pearson Correlation (R) | ||||||
---|---|---|---|---|---|---|

Sentinel-1 | Soil Moisture GS3 | Number of Observations | ||||

Polarization | Track | Orbit Pass | 5 cm | 10 cm | 20 cm | N |

VH | 153 | D ^{1} | 0.49 | 0.34 | 0.40 | 57 |

29 | A ^{2} | 0.51 | 0.39 | 0.49 | 70 | |

131 | A ^{2} | 0.56 | 0.46 | 0.59 | 66 | |

VV | 153 | D ^{1} | 0.47 | 0.27 | 0.36 | 57 |

29 | A ^{2} | 0.40 | 0.22 | 0.28 | 70 | |

131 | A ^{2} | 0.55 | 0.39 | 0.52 | 66 |

^{1}Descending,

^{2}Ascending.

**Table 3.**Pearson’s correlation (R values) for the grassland site between σ° VH and VV from S-1, and soil moisture (GS3) measured in situ at three depths: 5, 10, and 20 cm.

Grassland 2015–2017 Pearson Correlation (R) | ||||||
---|---|---|---|---|---|---|

Sentinel-1 | Soil Moisture GS3 | Number of Observations | ||||

Polarization | Track | Orbit Pass | 5 cm | 10 cm | 20 cm | N |

VH | 153 | D ^{1} | 0.48 | 0.48 | 0.48 | 67 |

29 | A ^{2} | 0.47 | 0.49 | 0.49 | 79 | |

80 | D ^{1} | 0.28 | 0.29 | 0.27 | 73 | |

131 | A ^{2} | 0.55 | 0.53 | 0.47 | 72 | |

VV | 153 | D ^{1} | 0.54 | 0.53 | 0.46 | 67 |

29 | A ^{2} | 0.58 | 0.58 | 0.50 | 79 | |

80 | D ^{1} | 0.39 | 0.37 | 0.26 | 73 | |

131 | A ^{2} | 0.72 | 0.69 | 0.55 | 72 |

^{1}Descending,

^{2}Ascending.

**Table 4.**Correlations between σ° VH and VV and SM at a 5 cm depth for the grassland and marshland sites during the seasons of 2015–2016.

Month | NDVI | SM and σ° VH | SM and σ° VV | N ^{4} | |||||
---|---|---|---|---|---|---|---|---|---|

Mean | SD ^{1} | R ^{2} | p-Value | S ^{3} dB/Vol.% | R ^{2} | p-Value | S ^{3} dB/Vol.% | ||

March | 0.43 | 0.08 | 0.87 | 0.00 | 0.24 | 0.64 | 0.02 | 0.14 | 13 |

April | 0.44 | 0.10 | 0.83 | 0.00 | 0.14 | 0.86 | 0.00 | 0.12 | 30 |

May | 0.63 | 0.16 | 0.87 | 0.00 | 0.10 | 0.85 | 0.00 | 0.11 | 33 |

June | 0.78 | 0.08 | 0.58 | 0.00 | 0.04 | 0.18 | 0.35 | - | 29 |

July | 0.77 | 0.06 | 0.61 | 0.00 | 0.04 | 0.19 | 0.27 | - | 31 |

August | 0.80 | 0.08 | 0.53 | 0.00 | 0.03 | 0.24 | 0.17 | - | 32 |

September | 0.76 | 0.07 | 0.83 | 0.00 | 0.06 | 0.51 | 0.00 | 0.03 | 35 |

October | 0.63 | 0.10 | 0.86 | 0.00 | 0.07 | 0.80 | 0.00 | 0.06 | 41 |

^{1}Standard deviations,

^{2}Correlation coefficient,

^{3}Sensitivity,

^{4}Number of observations.

**Table 5.**Kendall’s tau statistics between the NDVI and the σ° indices VH−VV and VV/VH for the grassland and marshland sites.

Site | Year | S-1 Track | Kendall’s Tau for VH−VV | N ^{1} | Kendall’s Tau for VV/VH |
---|---|---|---|---|---|

grassland | 2016 | 29 | 0.52 | 37 | 0.42 |

131 | 0.39 | 36 | 0.46 | ||

2017 | 29 | 0.51 | 27 | 0.28 | |

131 | 0.50 | 26 | 0.28 | ||

marshland | 2016 | 29 | 0.35 | 37 | 0.37 |

131 | 0.54 | 36 | 0.56 | ||

2017 | 29 | 0.68 | 27 | 0.39 | |

131 | 0.74 | 25 | 0.32 |

^{1}Number of observations.

**Table 6.**Mean absolute percentage error (MAPE) errors of σ° VH and VV derived from Model 1a and Model 1b for the years 2015–2017.

Site | Track | MAPE1 ^{1} (%) | MAPE2 ^{2} (%) | Number of Observations |
---|---|---|---|---|

Grassland | 131 | 5.7 | 8.7 | 62 |

29 | 5.9 | 8.8 | 56 | |

Marshland | 131 | 7.6 | 8.8 | 45 |

29 | 7.2 | 8.8 | 47 | |

All | 6.6 | 8.8 | 200 |

^{1}Errors applies to Model 1a,

^{2}Errors applies to Model 1b.

SM Range | N ^{1} | RMSE (vol. %) |
---|---|---|

20–40 | 39 | 11.8 |

40–60 | 39 | 9.5 |

60–80 | 35 | 9.9 |

80–100 | 34 | 8.4 |

All | 147 | 9.8 |

^{1}Number of observations.

NDVI Range | N ^{1} | RMSE (vol. %) |
---|---|---|

0.2–0.4 | 24 | 7.4 |

0.4–0.6 | 40 | 8.5 |

0.6–0.8 | 47 | 10.4 |

0.8–0.9 | 36 | 11.5 |

All | 147 | 9.8 |

^{1}Number of observations.

**Table 9.**Errors analysis for different ranges of SM (5 cm depth) as retrieved by Model 2 (whole year).

SM Range (vol. %) | N ^{1} | RMSE (vol. %) |
---|---|---|

20–40 | 51 | 14.8 |

40–60 | 53 | 11.8 |

60–80 | 64 | 13.8 |

80–100 | 40 | 13.6 |

All | 252 | 13.0 |

^{1}Number of observations.

**Table 10.**Errors analysis for different ranges of SM (5 cm depth) as retrieved by Model 2 (growing season).

SM Range (vol. %) | N ^{1} | RMSE (vol. %) |
---|---|---|

20–40 | 60 | 12.1 |

40–60 | 70 | 12.3 |

60–80 | 60 | 14.7 |

80–100 | 62 | 11.8 |

All | 208 | 13.5 |

^{1}Number of observations.

**Table 11.**Errors analysis for different ranges of SM (5 cm depth) retrieved by Model 2 for validation data.

SM Range (vol. %) | N ^{1} | RMSE (vol. %) |
---|---|---|

40–60 | 14 | 11.0 |

60–80 | 29 | 8.5 |

80–100 | 35 | 15.2 |

All | 76 | 12.6 |

^{1}Number of observations.

Orbit/Track | Model 1a | Model 1b | Model 2 |
---|---|---|---|

RMSE (vol. %) | |||

A/29 | 10.0 | 10.8 | 15.2 |

A/131 | 9.5 l | 9.9 | 10.7 |

© 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## Share and Cite

**MDPI and ACS Style**

Dabrowska-Zielinska, K.; Musial, J.; Malinska, A.; Budzynska, M.; Gurdak, R.; Kiryla, W.; Bartold, M.; Grzybowski, P. Soil Moisture in the Biebrza Wetlands Retrieved from Sentinel-1 Imagery. *Remote Sens.* **2018**, *10*, 1979.
https://doi.org/10.3390/rs10121979

**AMA Style**

Dabrowska-Zielinska K, Musial J, Malinska A, Budzynska M, Gurdak R, Kiryla W, Bartold M, Grzybowski P. Soil Moisture in the Biebrza Wetlands Retrieved from Sentinel-1 Imagery. *Remote Sensing*. 2018; 10(12):1979.
https://doi.org/10.3390/rs10121979

**Chicago/Turabian Style**

Dabrowska-Zielinska, Katarzyna, Jan Musial, Alicja Malinska, Maria Budzynska, Radoslaw Gurdak, Wojciech Kiryla, Maciej Bartold, and Patryk Grzybowski. 2018. "Soil Moisture in the Biebrza Wetlands Retrieved from Sentinel-1 Imagery" *Remote Sensing* 10, no. 12: 1979.
https://doi.org/10.3390/rs10121979