Study On Microwave Signal Penetration In Finland Boreal Forest Based On Airborne Profiling Radar

Forest vertical structure is used to describe the forest dynamic change from canopy to bottom.It is derived from the selection of natural conditions such as sunlight,moisture,etc.,which affects the growth of forest and the succession.Forest vertical structure has significant meaning for the accuracy of forest vertical biodiversity estimation.Furthermore,it plays an important role in the process of forest succession,net primary productivity,biodiversity,carbon cycle and global change.Small footprint airborne profile radar has great potential in the detection of forest vertical structures.It has great penetration in vegetation due to its longer wavelength comparing with optical active and passive sensors.In addition,its range resolution supports the estimation of forest parameters in complex forest spatial structures,compare with passive remote sensing and LiDAR data.The quantitative study of the penetration depth of forest canopy by microwave radar can provide clues for forest scattering mechanism and help to improve the accuracy of surface elevation and canopy height extraction,thus providing a more accurate and efficient means for forest vertical structure detection.The forest leaf area index and the canopy height are the important parameters for canopy density and biomass estimation,which directly affect the accuracy of the biomass model.The article revolves around the aforementioned content,based on the airborne field test of the northern forest of southern Finland,and using the frequency modulated continuous wave(FMCW)radar and the airborne discrete laser radar(LIDAR)to collect data simultaneously to carry out the following research work:(1)According to the characteristics of echo waveforms of forest targets,a microwave waveform preprocessing algorithm is established.The Fast Fourier Transform(FFT)transform operation is used to process the collected raw data for obtaining the corresponding signal spectrum,and its amplitude spectrum distribution is solved.Based on the conversion relationship between the microwave radar signal and the target scattering power signal,the signal amplitude spectrum is converted to original power waveform signal.The original power waveform data in the filter window range is convoluted with the filter weights to achieve the fitting of the original power waveform,and the echo characteristic parameters are extracted.The ground height and canopy height extracted by microwave radar are compared with the corresponding heights extracted by the LIDAR point cloud.The quantitative analysis results show that the microwave radar signal has advantages over LIDAR in ground elevation measurement.Especially in density forest,the advantage of microwave radar signal penetration is more obvious,and it can obtain accurate surface elevation.(2)Quantitative analysis of microwave signals penetration in forest.Different spatial distribution types in the vertical direction of the forest is analyzed.The accumulated energy curve obtained by integrating the vertical scattering energy indicates that the microwave radar energy saturation point appears deeper than LiDAR in densely clustered and densely distributed forest stands.Microwave radar is more sensitive to the distribution of internal branches and leaves of canopy than LiDAR.The height of the energy curve quantile can quantitatively reflect the difference in the vertical penetration depth between the microwave radar and the LiDAR.The difference in radar estimated altitude values gradually increases,and the average height of LiDAR at the 90%cumulative energy is 1.37 meters higher than that of microwave radars.The cumulative energy of the LiDAR h90 is discrete in the vertical direction,with a maximum of 18m,while that of microwave radar concentrates at 4-5m.It presents the fact that the backscattering of LiDAR is happened in the top of crown,the backscattering in the lower canopy is weak due to its poor penetration capability.It means the microwave radar has a deeper detection depth in the forest,and it can acquire the scattered signals of the dense ground canopy.(3)Calculation of forest LAI by airborne profile microwave radar with different polarization data.In this paper,based on Beer-Lambert law,the leaf area index of forest using LiDAR point cloud is used as a reference value.Results shows that the best scale of LiDAR matched with the microwave radar is the main beamwidth in terms of LAI estimation.The energy amplitude of forest profiles obtained by different polarization methods show that the cross-polarized mainly come from the canopy,while the co-polarized comes from the ground,and the forest canopy severely degrades the cross-polarization.With LAI extracted from the cross-polarization waveform and LiDAR point cloud,R2 is 0.60 and co-polarization is 0.65.It shows that the airborne profiling microwave radar has certain potential in the extraction of forest leaf area index.