The Planetary Variables Biomass Proxy is based on a fusion of microwave and optical satellite imagery. It uses the vegetation signal as derived from cross polarizations of microwave observations in combination with the vegetation indicators from the optical Sentinel 2 imagery. The vegetation signal in the microwave region is different as compared to in the optical domain. It is a direct function of the vegetation water content, dielectric properties of the water within the vegetation, and the vegetation structure (e.g. see Ulaby et al., 1986; Kerr et al., 1994). Considering the strong relationship between vegetation water content and vegetation biomass, this microwave information is, in this case, used as a biomass indicator for crops (e.g. see Vreugdenhil et al., 2018; Khabbazan et al., 2019). On the other hand, in optical imagery, spectral conditions are used to estimate the vegetation conditions. The spectral measurements in the visible region are sensitive to the chlorophyll content, while the measurements in the near infrared are sensitive to the mesophyll structure of the leaves (Townshend, 1993).


The Biomass Proxy algorithm is designed for agricultural crops and uses both signals. For the development of this index the following steps were taken:

  1. Pre-process the optical and microwave data using multiple cloud masking routines for the optical imagery and multi temporal filtering for the microwave images, as well as orbit corrections and bare soil mitigation from multi polarization derivatives. Preprocessing of optical data involves cloud masking the Sentinel-2 images by combining the Fmask (Zhu et al. 2015), S2cloudless (Sanchez et al. 2020), and Sen2Cor (Zhu et al. 2015) algorithms and snow and water masking by combining Fmask and Sen2cor. The masked pixels are assigned a NoData value equal to 65535.

  2. Define a relationship between the different vegetation signals derived from both the optical and microwave domain.

  3. For each field, build the temporal vegetation signal based on a dynamic weighting of each signal.

  4. Redistribute for each day the derived temporal signal on each field pixel based on a different dynamic spatial weighting, generating the biomass maps.


Optical vegetation signal (NDVI)

NDVI is an abbreviation for Normalized Difference Vegetation Index and it is derived from optical satellite imagery. It is the most widely used vegetation index that is derived from remote sensing. NDVI is a measure of chlorophyll activity in plants, which is responsible for the photosynthetic energy conversion in plants. Chlorophyll activity is highest in green leaves, therefore, generally speaking, the more green leaves a plant has, the higher the NDVI. Chlorophyll strongly absorbs visible light (from 0.4 to 0.7 µm) for use in photosynthesis, whereas plant cells strongly reflect near-infra-red light (0.7 to 1.1 µm). Since optical satellites generally measure bands in both spectra, an index can be constructed that makes use of the difference to evaluate chlorophyll activity. To make the index comparable worldwide it is normalized. The resulting index is NDVI:

\[\frac{\mathrm{NIR} - \mathrm{RED}}{\mathrm{NIR} + \mathrm{RED}}\]

Where NIR is the reflection in the near-infrared spectrum and RED is the reflection in the red (visible) range of the spectrum.

By design, the NDVI itself thus varies between -1.0 and +1.0. The higher the value, the more chlorophyll activity and therefore the more abundant, and the healthier the vegetation.

Optical data is collected at a resolution of 10x10m from the Sentinel-2 satellites.

Microwave signal

Active microwaves are in an entirely different part of the spectrum than optical waves but are also known for their ability to detect changes in vegetation. Since VV and VH backscatter - two different polarizations of SAR (synthetic aperture radar) - respond differently to vegetation, information on vegetation structure, vegetation water content, and the dielectric properties of vegetation water can be obtained (e.g. see Ulaby et al., 1986; Kerr et al., 1994). There are other factors which influence the backscatter signal such as moisture content and the texture and roughness of the underlying soil, so it is therefore important to isolate the vegetation signal in the radar signals.

It is more challenging to isolate the vegetation signal, than with optical signals, however when successful, active microwaves are a very consistent and reliable source of information. This is because, unlike optical signals, they penetrate cloud cover.

Microwave data is collected at a resolution of 10x10m from the Sentinel-1 satellites.


Kerr, Y.H. and Wigneron, J.-P. 1994: Vegetation models and observations - a review. In Choudhury, B.J. , Kerr, Y.H. , Njoku, E.G. and Pampaloni, P. , editors, Passive microwave remote sensing research related to land-atmosphere interactions, Utrecht: VSP, 317-44.

Khabbazan, S., Vermunt, P., Steele-Dunne, S., Ratering Arntz, L., Marinetti, C., van der Valk, D., Iannini, L., Molijn, R., Westerdijk, K., & van der Sande, C. (2019). Crop monitoring using Sentinel-1 data: A case study from The Netherlands. Remote Sensing, 11(16), 1887.

Main-Knorn, M.; Pflug, B.; Louis, J.; Debaecker, V.; Müller-Wilm, U.; Gascon, F. Sen2Cor for Sentinel-2. In Proceedings of the Image and Signal Processing for Remote Sensing XXIII, Warsaw, Poland, 11–14 September 2017; Bruzzone, L., Bovolo, F., Benediktsson, J.A., Eds.; SPIE: Bellingham,WA, USA, 2017; p. 3.

Michel, J., Youssefi, D., & Grizonnet, M. (2014). Stable mean-shift algorithm and its application to the segmentation of arbitrarily large remote sensing images. IEEE Transactions on Geoscience and Remote Sensing, 53(2), 952-964.

Sanchez, A.H.; Picoli, M.C.A.; Camara, G.; Andrade, P.R.; Chaves, M.E.D.; Lechler, S.; Soares, A.R.; Marujo, R.F.B.;, R.E.O.; Ferreira, K.R. et al. 2020. Comparison of Cloud Cover Detection Algorithms on Sentinel–2 Images of the Amazon Tropical Forest. Remote Sensing 12, 1284.

Townshend, J. R. G., Tucker, C. J., and Goward, S. N. (1993). Global vegetation mapping. In Gurney, R. J., Foster, J. L., and Parkinson, C. L., editors, Atlas of Satellite Observations related to Global Change. Cambridge University Press, New York.

Ulaby, F. T., Moore, R. K., & Fung, A. K. (1986). Microwave remote sensing: Active and passive. Volume 3-From theory to applications.

Valero, S., Morin, D., Inglada, J., Sepulcre, G., Arias, M., Hagolle, O., Dedieu, G., Bontemps, S., Defourny, P., & Koetz, B. (2016). Production of a dynamic cropland mask by processing remote sensing image series at high temporal and spatial resolutions. Remote Sensing, 8(1), 55.

Vreugdenhil, M., Wagner, W., Bauer-Marschallinger, B., Pfeil, I., Teubner, I., Rüdiger, C., & Strauss, P. (2018). Sensitivity of Sentinel-1 backscatter to vegetation dynamics: An Austrian case study. Remote Sensing, 10(9), 1396.

Zhu, Z.; Wang, S.; Woodcock, C.E. 2015. Improvement and expansion of the Fmask algorithm: Cloud, cloud shadow, and snow detection for Landsats 4–7, 8, and Sentinel 2 images. Remote Sens. Environ. 159, 269–277.