Application of Repeat-Pass TerraSAR-X Staring Spotlight Interferometric Coherence to Monitor Pasture Biophysical Parameters: Limitations and Sensitivity Analysis

• Published in: IEEE journal of selected topics in applied earth observations and remote sensing Vol. 10; no. 7; pp. 3225 - 3231
• Main Authors: Ali, Iftikhar, Barrett, Brian, Cawkwell, Fiona, Green, Stuart, Dwyer, Edward, Neumann, Maxim
• Format: Journal Article
• Language: English
• Published: IEEE 01.07.2017
• Subjects: Backscatter; Biomass; Biophysical parameters; Coherence; Decorrelation; InSAR coherence; Interferometry; managed pastures; Synthetic aperture radar; TerraSAR-X Staring Spotlight (TSX-ST); Wind speed
• ISSN: 1939-1404; 2151-1535
• DOI: 10.1109/JSTARS.2017.2679761

This paper describes the potential and limitations of repeat-pass synthetic aperture radar interferometry (InSAR) to retrieve the biophysical parameters of intensively managed pastures. We used a time series of eight acquisitions from the TerraSAR-X Staring Spotlight (TSX-ST) mode. The ST mode is different from conventional Strip map mode; therefore, we adjusted the Doppler phase correction for interferometric processing. We analyzed the three interferometric pairs with an 11-day temporal baseline, and among these three pairs found only one gives a high coherence. The results show that the high coherence in different paddocks is due to the cutting of the grass in the month of June, however the temporal decorrelation in other paddocks is mainly due to the grass growth and high sensitivity of the X-band SAR signals to the vegetation cover. The InSAR coherence (over coherent paddocks) shows a good correlation with SAR backscatter (R 2 dB = 0.65, p <; 0.05) and grassland biophysical parameters (R 2 Height = 0.55, p <; 0.05, R 2 Biom ass = 0.75, p <; 0.05). It is thus possible to detect different management practices (e.g., grazing, mowing/cutting) using SAR backscatter (dB) and coherence information from high spatial short baseline X-band imagery; however, the rate of decorrelation over vegetated areas is high. Initial findings from the June pair show the possibility of change detection due to the grass growth, grazing, and mowing events by using InSAR coherence information. However, it is not possible to automatically categorize different paddocks undergoing these changes based only on the SAR backscatter and coherence values, due to the ambiguity caused by tall grass flattened by the wind.