Extending the Ability of Near‐Infrared Images to Monitor Small River Discharge on the Northeastern Tibetan Plateau

Abstract

River discharge gauging is scarce in high mountainous regions, especially on the Tibetan Plateau, where rivers are widely distributed. Although remote sensing is an important mean of monitoring river discharge, previous methods are most suitable for large rivers, and the ability to monitor small rivers (with widths less than 100 m) is limited. To resolve this issue, a multiple pixel ratio (MPR) method is presented for monitoring the discharge of small rivers based on the relationship between river discharge and the near‐infrared reflectivity. Utilizing 281 Landsat images (1990–2015), we monitored river discharges in two subbasins in the upstream region of the Heihe River located on the northeastern Tibetan Plateau. Our results indicate that the performance of the MPR method is more stable than the previous calibration/measurement method. The monitoring accuracy was correlated with the length and location of the selected inundated river channel. Using selected inundated river channel lengths between 300 and 600 m can provide better monitoring accuracy. The Nash‐Sutcliffe efficiency of monitoring results (2013–2014) of Qilian station was 0.82; the Zhamashike station (2015) was 0.45; the two stations multiyears (1990–2013) monitoring results were 0.32 and 0.41; the ungauged basin was 0.45. Our results suggest that the MPR method can expand the ability of remote sensing to monitor discharge in small rivers with widths greater than 30 m on the Tibetan Plateau. In addition, the new method also has the potential to monitor the discharge of ungauged small rivers (with widths greater than 30 m).

River Flow Monitoring by Sentinel-3 OLCI and MODIS: Comparison and Combination

“…Li et al [21], considering the Landsat imagery at 30 m, preferred to resample the original satellite scenes in tiles of 3 × 3 pixels. Indeed, by averaging the reflectance in multiple pixels, the values of the M pixel are expected to be more stable and less affected by the geometric misalignment of the images comprising the temporal series.…”

“…This principle was already applied in several studies [16][17][18][19][20][21][22][23] and extended to the Terra/Aqua Moderate Resolution Imaging Spectroradiometer (MODIS), ENVISAT MEdium Resolution Imaging Spectrometer (MERIS) and Landsat datasets. Specifically, in accordance with study by Tarpanelli et al [19], MODIS resulted in being the best performer, especially thanks to its daily temporal resolution and ease of data processing, two features not shared by MERIS and Landsat.…”

“…Specifically, in accordance with study by Tarpanelli et al [19], MODIS resulted in being the best performer, especially thanks to its daily temporal resolution and ease of data processing, two features not shared by MERIS and Landsat. However, the MODIS sensor is not accurate for small rivers (<100 m width) and to compensate for this limitation, the studies of Li et al [21] and Shi et al [22] investigated the use of a high-resolution satellite. Li et al [21] used Landsat in three rivers on the Tibet Plateau ranging from 60 m to 130 m width, and obtained Nash-Sutcliffe efficiency, NSE, ranging from 0.32 to 0.82.…”

“…However, the MODIS sensor is not accurate for small rivers (<100 m width) and to compensate for this limitation, the studies of Li et al [21] and Shi et al [22] investigated the use of a high-resolution satellite. Li et al [21] used Landsat in three rivers on the Tibet Plateau ranging from 60 m to 130 m width, and obtained Nash-Sutcliffe efficiency, NSE, ranging from 0.32 to 0.82. Shi et al [22] faced the estimation of river discharge at two study areas in the Murray-Darling basin with 60-90 m width analyzing the Harmonized Landsat and Sentinel-2 surface reflectance product [24], and obtained relative root mean square error, rRMSE, ranging from 19% and 53%.…”

Discharge Estimation Using Integrated Satellite Data and Hybrid Model in the Midstream Yangtze River

“…Alternatively, as one of the most widely used optical satellite data, the near infrared (NIR) band of Moderate Resolution Imaging Spectroradiometer (MODIS) sensors can identify the grid cells within and outside the river by distinctive reflectance values, and their ratio was also exploited to estimate discharge in medium-sized catchments (area < 10,000 km 2 ) [13,14]. Landsat data were also utilized to generate discharge because of fine resolution (30 m) and long duration (~50 years up to present) [15]. Considering the long visit period of 16 days of Landsat series satellites, the two Sentinel-2 satellites launched in 2015 and 2017 have managed to provide high-quality optical images with spatial resolution ranging from 10 to 60 m and a cycle period between 3 and 5 days [16,17].…”