Qing Tang

Due to fragmentized terrain and physiognomy of typical loess landform, long-term anthropogenic influences, and inherent vulnerability, soil erosion is a serious problem in the Loess Plateau of China. There is a critical need to assess soil erosion and spatial distribution for achieving sustainable land use and comprehensive soil conservation management. Taking the Yangou watershed as a case and using the Landsat Thematic Mapper image (land use map), Digital Elevation Model (DEM), soil maps, and precipitation data, this study integrated the revised universal soil loss equation (RUSLE) with GIS technology to estimate soil loss and its spatial distribution. The benefits of soil conservation of land use types were analyzed and the measures for future soil conservation planning were discussed. The results show that silt-covered land and terrace have high benefits of soil conservation, indicating that building check dam, producing silt-covered land for farming, and converting sloped farmland to terrace are effective ways to control soil erosion in the Yangou watershed. Furthermore, increasing vegetation coverage on lower coverage grassland, especially on the slopes with gradients >25°, and eliminating human disturbance in barren areas are feasible and effective measures for soil conservation planning. This study reveals that the integrated RUSLE-GIS model can evaluate and map soil erosion quantitatively and spatially at watershed scale in the Loess Plateau of China. The findings suggest strategies for coping with future soil conservation planning and provide valuable references for future assessments both in the Loess Plateau of China and elsewhere.

Contemporary cochlear implants with multiple electrode stimulation can produce good speech perception but poor music perception. Hindered by the lack of a gold standard to quantify electric pitch, relatively little is known about the nature and extent of the electric pitch abnormalities and their impact on cochlear implant performance. Here we overcame this obstacle by comparing acoustic and electric pitch perception in 3 unilateral cochlear-implant subjects who had functionally usable acoustic hearing throughout the audiometric frequency range in the non-implant ear. First, to establish a baseline, we measured and found slightly impaired pure tone frequency discrimination and nearly perfect melody recognition in all 3 subjects' acoustic ear. Second, using pure tones in the acoustic ear to match electric pitch induced by an intra-cochlear electrode, we found that the frequency-electrode function was not only 1-2 octaves lower, but also 2 times more compressed in frequency range than the normal cochlear frequency-place function. Third, we derived frequency difference limens in electric pitch and found that the equivalent electric frequency discrimination was 24 times worse than normal-hearing controls. These 3 abnormalities are likely a result of a combination of broad electric field, distant intra-cochlear electrode placement, and non-uniform spiral ganglion cell distribution and survival, all of which are inherent to the electrode-nerve interface in contemporary cochlear implants. Previous studies emphasized on the "mean" shape of the frequency-electrode function, but the present study indicates that the large "variance" of this function, reflecting poor electric pitch discriminability, is the main factor limiting contemporary cochlear implant performance.