The resolution of ocean reanalysis datasets is generally low because of the limited resolution of their associated numerical models.Low-resolution ocean reanalysis datasets are therefore usually interpolated to provid...The resolution of ocean reanalysis datasets is generally low because of the limited resolution of their associated numerical models.Low-resolution ocean reanalysis datasets are therefore usually interpolated to provide an initial or boundary field for higher-resolution regional ocean models.However,traditional interpolation methods(nearest neighbor interpolation,bilinear interpolation,and bicubic interpolation)lack physical constraints and can generate significant errors at land-sea boundaries and around islands.In this paper,a machine learning method is used to design an interpolation algorithm based on Gaussian process regression.The method uses a multiscale kernel function to process two-dimensional space meteorological ocean processes and introduces multiscale physical feature information(sea surface wind stress,sea surface heat flux,and ocean current velocity).This greatly improves the spatial resolution of ocean features and the interpolation accuracy.The eff ectiveness of the algorithm was validated through interpolation experiments relating to sea surface temperature(SST).The root mean square error(RMSE)of the interpolation algorithm was 38.9%,43.7%,and 62.4%lower than that of bilinear interpolation,bicubic interpolation,and nearest neighbor interpolation,respectively.The interpolation accuracy was also significantly better in off shore area and around islands.The algorithm has an acceptable runtime cost and good temporal and spatial generalizability.展开更多
Gully erosion is a worldwide problem of land degradation and water quality,and it is also frequent in Brazil.Typically,anthropic influence is the major driver of gully evolution.To study and monitor gullies it is nece...Gully erosion is a worldwide problem of land degradation and water quality,and it is also frequent in Brazil.Typically,anthropic influence is the major driver of gully evolution.To study and monitor gullies it is necessary to use specific instruments and methods to obtain accurate information.The objective of this study was to use Terrestrial Laser Scanning(TLS) to create digital elevation model(DEM) accurately and define morphometric variables that characterize gullies in a mountainous relief.Two different interpolations were evaluated using the Topogrid and GridSurfaceCreate algorithms to elaborate DEM.Topographic profile for gullies was used to assess modeling quality.The DEM of the Gully 1(G1) from the Topogrid algorithm estimated soil loss of 49%,whereas the GridSurfaceCreate algorithm estimated a soil loss of97%,in a period of 1 year.The estimated soil loss for the Gully 2(G2) was 14% from the Topogrid,and 8%from the GridSurfaceCreate algorithm.The GridSurfaceCreate algorithm underestimated the volume to area ratio for G2 due to a failure on interpolating a region of low point representativity.The Topogrid algorithm represented better the terrain irregularities,as observed through the topographic profiles traced in three regions of G1 and G2.Statistical analysis showed that the GridSurfaceCreate algorithm presented lower accuracy in estimating elevations.The underestimation trend of this algorithm was also observed in G2.The gullies showed considerable soil losses,which may reduce the areas suitable for agricultural activities,and silting up of water courses.The Topogrid algorithm presented satisfactory results,denoting great potential to produce morphometric data of gullies.展开更多
基金Supported by the National Natural Science Foundation of China(Nos.41675097,41375113)。
文摘The resolution of ocean reanalysis datasets is generally low because of the limited resolution of their associated numerical models.Low-resolution ocean reanalysis datasets are therefore usually interpolated to provide an initial or boundary field for higher-resolution regional ocean models.However,traditional interpolation methods(nearest neighbor interpolation,bilinear interpolation,and bicubic interpolation)lack physical constraints and can generate significant errors at land-sea boundaries and around islands.In this paper,a machine learning method is used to design an interpolation algorithm based on Gaussian process regression.The method uses a multiscale kernel function to process two-dimensional space meteorological ocean processes and introduces multiscale physical feature information(sea surface wind stress,sea surface heat flux,and ocean current velocity).This greatly improves the spatial resolution of ocean features and the interpolation accuracy.The eff ectiveness of the algorithm was validated through interpolation experiments relating to sea surface temperature(SST).The root mean square error(RMSE)of the interpolation algorithm was 38.9%,43.7%,and 62.4%lower than that of bilinear interpolation,bicubic interpolation,and nearest neighbor interpolation,respectively.The interpolation accuracy was also significantly better in off shore area and around islands.The algorithm has an acceptable runtime cost and good temporal and spatial generalizability.
基金the FAPERJ for the concession scholarships for the first author (Grants No. E26/101.897/2010 - 63010)funded by the Pró-Equipamentos program for Capes (Coordenacao de Aperfeicoamento de Pessoal de Nível Superior)。
文摘Gully erosion is a worldwide problem of land degradation and water quality,and it is also frequent in Brazil.Typically,anthropic influence is the major driver of gully evolution.To study and monitor gullies it is necessary to use specific instruments and methods to obtain accurate information.The objective of this study was to use Terrestrial Laser Scanning(TLS) to create digital elevation model(DEM) accurately and define morphometric variables that characterize gullies in a mountainous relief.Two different interpolations were evaluated using the Topogrid and GridSurfaceCreate algorithms to elaborate DEM.Topographic profile for gullies was used to assess modeling quality.The DEM of the Gully 1(G1) from the Topogrid algorithm estimated soil loss of 49%,whereas the GridSurfaceCreate algorithm estimated a soil loss of97%,in a period of 1 year.The estimated soil loss for the Gully 2(G2) was 14% from the Topogrid,and 8%from the GridSurfaceCreate algorithm.The GridSurfaceCreate algorithm underestimated the volume to area ratio for G2 due to a failure on interpolating a region of low point representativity.The Topogrid algorithm represented better the terrain irregularities,as observed through the topographic profiles traced in three regions of G1 and G2.Statistical analysis showed that the GridSurfaceCreate algorithm presented lower accuracy in estimating elevations.The underestimation trend of this algorithm was also observed in G2.The gullies showed considerable soil losses,which may reduce the areas suitable for agricultural activities,and silting up of water courses.The Topogrid algorithm presented satisfactory results,denoting great potential to produce morphometric data of gullies.
