the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 29 Jan 2026
High-resolution monthly glacier surface velocity mapping in the Kangri Karpo region (2015–2024) using multi-source remote sensing data fusion
Abstract. To improve the accuracy and timeliness of glacier surface-velocity retrieval in complex mountain terrain, we develop a high-resolution fusion method combining Landsat, Sentinel-1/2, and UAV (Unmanned Aerial Vehicle) data, and produce monthly velocity products for the Kangri Karpo region for 2015–2024. Compared with existing large-area public datasets, the products offer markedly higher spatial resolution and better detection of small mountain glaciers; relative to single-sensor inputs prior to fusion, the valid-pixel ratio increases by ~50 %, the average number of valid months per pixel over the decade rises by ~50, and spatial smoothness improves—demonstrating the method’s suitability for rugged terrain. Spatially, velocities follow the canonical “fast center, slow margins” pattern, with multi-year maxima >700 m·yr⁻¹ and values in lower reaches and most tributaries generally <100 m·yr⁻¹. Attribute analysis indicates significant correlations between velocity and area, slope, and aspect: larger glaciers flow faster overall; within individual glaciers, velocity responds more strongly to slope; after controlling for area and slope, south-facing glaciers are slightly faster. Temporally, the intra-annual series shows clear seasonality, with peaks at the beginning and end of the melt season and sustained high speeds throughout. At the interannual scale, most pixelwise decadal trends lie within −0.1 to +0.1 m·d⁻¹·dec⁻¹ (overall subdued change), and the median trend is slightly positive, indicating weak regional acceleration; ~38.3 % of glaciers accelerate significantly, 25.5 % decelerate significantly, and 36.2 % show no significant trend (p ≥ 0.05). By aspect, significant acceleration is concentrated on south- and west-facing glaciers, whereas significant deceleration occurs mainly on east- and north-facing glaciers. Month-resolved trends indicate acceleration primarily in April–May (~0.15–0.20 m·d⁻¹·dec⁻¹), likely linked to enhanced meltwater input from an advanced melt season, and deceleration concentrated in July–August (≤ −0.15 m·d⁻¹·dec⁻¹), plausibly associated with intensified mass deficit.
Competing interests: At least one of the (co-)authors is a member of the editorial board of The Cryosphere.
Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this paper. While Copernicus Publications makes every effort to include appropriate place names, the final responsibility lies with the authors. Views expressed in the text are those of the authors and do not necessarily reflect the views of the publisher.- Preprint
(2774 KB) - Metadata XML
-
Supplement
(2257 KB) - BibTeX
- EndNote
Status: open (until 20 Mar 2026)
-
RC1: 'Comment on egusphere-2025-6357', Anonymous Referee #1, 22 Feb 2026
reply
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2026/egusphere-2025-6357/egusphere-2025-6357-RC1-supplement.pdfReplyCitation: https://doi.org/
10.5194/egusphere-2025-6357-RC1 -
AC1: 'Reply on RC1', Daoxun Gao, 23 Feb 2026
reply
回复审稿人#1：
我们感谢审稿人提出的详细意见，并已根据所有建议进行了修改。在本回复中，审稿人的意见以黑色标准字体显示。我们的回复以蓝色标准字体显示，对稿件的修改以蓝色粗体显示。
主要评论：
1）在第3.2节中，无人机覆盖范围似乎仅限于研究区域内单个冰川的一小部分末端区域。请阐明该训练区域在空间和时间（季节性/全年条件）上的代表性程度，并论证将无人机校准的参数/权重外推至整个区域和完整年度周期时的适用性。
感谢您的评论。我们已在3.2节中阐明，尽管无人机覆盖范围仅限于亚农冰川末端附近约30平方公里的区域，但无人机获取的数据涵盖了以下几个方面：(i) 从消融季到早期积累期（2023年6月至11月）的季节性条件；(ii) 从快速流动的冰川干流到流动较慢的边缘区域，呈现出清晰的空间速度梯度；(iii) 包括碎屑覆盖冰、洁净冰和裂缝带在内的多种表面相。这些特征提供了异质但相关的条件，使得无人机数据子集具有潜在的通用性，可用于评估Landsat、Sentinel-1和Sentinel-2速度反演的性能，以及校准本研究中使用的融合权重。
此外，我们使用配备M6 Pro (M6P)测绘相机的DJI Matrice 300 RTK无人机获取了用于参考和评估的无人机摄影测量数据。2023年6月至11月期间，我们采集了六幅正射影像，构成五组影像对。虽然覆盖范围仅限于亚农冰川末端附近约30平方公里的区域，但无人机测量涵盖了多种不同的冰川条件：(i) 测量时间段涵盖了消融期到早期积累期的过渡阶段；(ii) 测绘区域既包括快速流动的冰川干流区域，也包括流速较慢的冰川边缘；(iii) 表面类型包括覆盖碎屑的冰、洁净冰和裂缝区域。因此，这组无人机数据在空间和时间上具有潜在的可迁移性，可用于评估Landsat、Sentinel-1和Sentinel-2冰川速度反演的性能，以及校准区域融合中的融合权重。
2) 虽然稿件中指出来自三个数据源的速度数据已统一调整至 30 米的空间分辨率，但不同传感器/产品的底层原始网格未必已配准。为了确保逐像素加权最小二乘法 (WLS) 拟合及后续融合的有效性和可重复性，应明确描述融合前的配准/统一步骤，包括参考网格的选择、重投影和重采样方法（以及插值方案）、网格对齐策略、无数据/掩码传播规则，以及这些操作是否在 WLS 拟合之前执行。
Thank you for this helpful suggestion. We agree that explicit description of the pre-fusion grid harmonization is necessary for ensuring the validity and reproducibility of the pixel-wise WLS fitting and subsequent fusion. We have therefore added a brief statement in Sect. 3.3 clarifying (i) the reference grid used for harmonization, (ii) the resampling method adopted for co-registration, and (iii) that these operations are performed prior to the WLS fitting.
“To support pixel-wise WLS fitting and the subsequent fusion, all velocity products were harmonized to a common 30 m reference grid prior to WLS. We used the Sentinel-2 COSI-Corr velocity image as the reference geometry, and co-registered the Landsat-, Sentinel-1-, and UAV-derived velocity image to this grid using nearest-neighbour resampling, so that the original velocity values are preserved without interpolation smoothing.”
3) For the WLS-based fusion weights, please specify the exact objective function, the interpretation ofthe weights within the WLS formulation, and any constraints imposed during optimization. In particular, clarify whether non-negativity and/or a normalization constraint are enforced. If constraints are used, briefly describe the corresponding solution strategy/implementation to ensure reproducibility.
Thank you for this comment. We have clarified the WLS formulation by explicitly stating the constraints imposed on the fusion weights. In our implementation, the weights are constrained to be non-negative and to sum to one.
“Let denote the UAV-derived velocity at pixel (, and ,, denote the Landsat, Sentinel-1, and Sentinel-2 velocities, with corresponding fusion weights ,,. The WLS objective is:
After defining the objective in Eq. (1), we estimate the fusion weights by minimizing subject to the constraints , , and . This simplex constraint ensures that the weights are directly interpretable as non-negative mixing fractions. The resulting optimal weight triplet is then applied to fuse the remaining monthly velocity maps.”
4) In Sect. 3.4, (𝑖,𝑗)is defined as the pixel location in the image. However, in Eq. (1) the weight estimation would normally iterate over all pixels of the 2-D grid, whereas the summation appears to run only over 𝑖. Sentinel-1–derived velocities are used as the information source for filling gaps in the fused product (via the enhancement-coefficient field), but the rationale for selecting Sentinel-1 as the gap-filling reference is not sufficiently articulated. Also, the first two paragraph of Section 3.4 is a bit repetitive with Section 3.1.
Thank you for these helpful comments. We have revised Sect. 3.4 accordingly.
(1) We adjusted Eq. (1) and its notation to clearly indicate that the WLS objective is summed over the set of overlapping valid pixels on the 2-D grid.
(2) Before introducing the enhancement-coefficient infilling, we added a short rationale explaining why Sentinel-1 is used to guide gap repair: optical (Landsat/Sentinel-2) velocities are often missing due to cloud/snow/illumination limitations, whereas Sentinel-1 provides a more complete auxiliary field, helping preserve glacier-motion patterns while improving continuity.
(3) We rewrote the first two paragraphs of Sect. 3.4 to reduce redundancy with Sect. 3.1 and to focus the section on the weight-estimation formulation and subsequent steps.
“In this section, we describe how fusion weights are estimated using co-temporal UAV-derived velocities. Monthly velocity maps from Landsat, Sentinel-1, and Sentinel-2 are generated independently, and UAV orthomosaics are processed to provide a high-accuracy reference for the same periods. A WLS formulation is then used to estimate the optimal weights for the three satellite products, which are applied in the subsequent fusion.
Let denote the UAV-derived velocity at pixel (, and ,, denote the Landsat, Sentinel-1, and Sentinel-2 velocities, with corresponding fusion weights ,,. The WLS objective is:
After defining the objective in Eq. (1), we estimate the fusion weights by minimizing subject to the constraints , , and . This simplex constraint ensures that the weights are directly interpretable as non-negative mixing fractions. The resulting optimal weight triplet is then applied to fuse the remaining monthly velocity maps.
In the fusion stage, to address pixels with missing values (NoData), we introduce a binary mask to locally renormalize the weights, so that the weighted average at ( is computed only over sources that are valid there (Landsat or Sentinel-2). The preliminary fused velocity is:
Although weighted fusion can effectively integrate multi-source information, some areas may still contain NoData. To further fill these gaps and enhance data continuity, this study introduces a sliding-window enhancement-coefficient infilling method. Because Landsat- and Sentinel-2–derived velocity maps often exhibit large gaps in this cloud- and snow-prone region and simple interpolation can be unreliable, we use Sentinel-1 SAR velocities—less sensitive to clouds and illumination—as a more complete auxiliary field to guide gap repair. Specifically, Sentinel-1 is used to infer the local spatial variation of the fused field and fill NoData accordingly: for each pixel , we consider a 10×10 sliding window (stride = 1 pixel) and estimate a local enhancement coefficient that describes the response of the fused value to Sentinel-1,:
”
5) In Eqs. (3) and (4), the enhancement coefficient is denoted as aΩ, which naturally reads as a single constant/parameter associated with a domain Ω. However, based on the stated computation and definition (window-based estimation followed by Gaussian smoothing to form a spatial field), the enhancement coefficient should vary spatially and thus be a gridded quantity.
Thank you for pointing this out. We agree that the enhancement coefficient is a spatially varying gridded field rather than a single constant associated with a window domain. We have therefore revised the notation in Sect. 3.4.
“Although weighted fusion can effectively integrate multi-source information, some areas may still contain NoData. To further fill these gaps and enhance data continuity, this study introduces a sliding-window enhancement-coefficient infilling method. Because Landsat- and Sentinel-2–derived velocity maps often exhibit large gaps in this cloud- and snow-prone region and simple interpolation can be unreliable, we use Sentinel-1 SAR velocities—less sensitive to clouds and illumination—as a more complete auxiliary field to guide gap repair. Specifically, Sentinel-1 is used to infer the local spatial variation of the fused field and fill NoData accordingly: for each pixel , we consider a 10×10 sliding window (stride = 1 pixel) and estimate a local enhancement coefficient that describes the response of the fused value to Sentinel-1,:
Only pixels where both and are valid are included in the summation. Next, is smoothed with a Gaussian filter to form a spatially varying enhancement-coefficient field over the entire image. Finally, for missing pixels in the fused map, infilling is performed as
Through this enhancement-based infilling, the spatial completeness of the fused image is markedly improved and gaps are reasonably reconstructed, providing more stable and continuous data support for subsequent glacier-change analyses and time-series modeling.”
6) In Sect. 4.1, the manuscript reports an “~70% / ~30% improvement” in valid pixels, but it is unclear whether this refers to an absolute increase (percentage points) or a relative increase with respect to a baseline. To avoid ambiguity, please provide the explicit definition/formula used to compute the improvement.
Thank you for this helpful comment. We agree that the phrase “~70% / ~30% improvement” is ambiguous. To avoid confusion, we have revised Sect. 4.1 to report the improvement in percentage points (pp) instead of relative percentages.
“Figure 3 presents the monthly fraction of valid pixels in velocity images from 2015–2024 for the two representative glaciers, Xirinongpu and Yanong. The results show that, in most months, the fused product attains a substantially higher valid-pixel ratio than either single-source dataset. Over Xirinongpu, the mean valid-pixel ratios for Landsat and Sentinel-2 are 57.4% and 54.7%, respectively, whereas the fused result reaches 97.9%, corresponding to absolute increases of 40.5 and 43.2 percentage points, respectively. Over Yanong, Landsat and Sentinel-2 achieve 74.7% and 77.1%, while the fused image further increases to 99.2%, corresponding to absolute increases of 24.5 and 22.1 percentage points, respectively. These findings indicate that the proposed fusion method effectively overcomes the limitations of individual sources and markedly enhances data availability and continuity.”
7) In Sect. 3.5, “image smoothness” is defined as the pixelwise standard deviation within the mask for each month (“…we compute the pixelwise standard deviation…”). However, later (Fig. 5 and associated text) the manuscript refers to using variance to assess image smoothness.
Thank you for pointing out this inconsistency. We agree that the terminology should be consistent. In our analysis of image smoothness, the metric used is variance, and the “standard deviation” wording in Sect. 3.5 was imprecise. We have revised Sect. 3.5 to consistently use variance when describing the image-smoothness assessment.
“(3) Image smoothness: For the three velocity products and for GoLIVE and ITS_LIVE, we compute the pixelwise variance within the mask for each month. Changes in variance indicate noise suppression and smoothing, enabling a comprehensive assessment of fusion-driven gains in coverage, temporal continuity, and smoothness.”
8) In Sect. 4.4, the manuscript states that the observed bimodal intra-annual velocity pattern “accords with” the subglacial hydrology-dynamics evolution framework for maritime glaciers. While this interpretation is plausible, the current discussion remains largely qualitative and lacks a clear evidential link to the data presented in this study.
Thank you for this important comment. We agree that the original wording could be interpreted as an overly strong attribution. In the revised manuscript, we have moderated the interpretation and reframed it as a plausible process-based explanation that is broadly consistent with the seasonal pattern observed in our data and with previous studies. We also added a citation to Nanni et al. (2023) to support that similar seasonal acceleration behaviour, including autumn acceleration, has been reported in mountain glaciers in other regions.
“To characterize intra-annual variability in the Kangri Karpo region, we extracted monthly velocities from the 2015–2024 fused product and averaged each calendar month across years to derive a climatological annual cycle (Fig. 11). The resulting series shows a clear seasonal signal: velocities are relatively low in January-March, increase from March onward, reach a first peak in May, remain comparatively stable during summer, rise again to a second peak in October, and then decline. This indicates a bimodal intra-annual velocity pattern in the study region.
This bimodal pattern is broadly consistent with process-based interpretations proposed for temperate/maritime glaciers, in which seasonal changes in subglacial drainage efficiency modulate basal sliding. In particular, early-season meltwater and rainfall inputs may enhance sliding under a relatively inefficient distributed drainage system, whereas subsequent drainage-system evolution toward more efficient channelized flow may reduce mean basal water pressure and limit velocities later in the melt season; partial re-closure or renewed pressurization may contribute to a later-season acceleration. Similar seasonal acceleration behaviour, including autumn acceleration, has also been reported for mountain glaciers in other regions (e.g., Nanni et al., 2023). However, we emphasize that the present study does not include direct observations of subglacial hydrology, and therefore this interpretation should be viewed as plausible rather than diagnostic.
The standard-deviation envelope in Fig. 11 further suggests month-dependent interannual variability. Standard deviations are larger in April-May, when seasonal acceleration initiates, indicating stronger year-to-year differences in the timing and magnitude of acceleration. By contrast, variability is comparatively smaller from June to the following March. Taken together, these results indicate a clear seasonal velocity response in the Kangri Karpo region, with a recurrent bimodal intra-annual structure.”
Nanni, U., Scherler, D., Ayoub, F., Millan, R., Herman, F., and Avouac, J.-P.: Climatic control on seasonal variations in mountain glacier surface velocity, The Cryosphere, 17, 1567–1583, https://doi.org/10.5194/tc-17-1567-2023, 2023.
9) Based on the velocity formulation described in Sect. 3.3, the primary unit of the derived glacier surface velocity should be m d-1. While, in several places (including Sect. 4.3) the manuscript reports velocities in m yr-1 or m/year. Please standardize the unit system throughout the manuscript.
Thank you for this helpful comment. We agree that the velocity unit should be standardized throughout the manuscript. The primary unit directly derived from the velocity formulation in Sect. 3.3 is m d⁻¹, and in the revised manuscript we have unified the velocity unit accordingly. Specifically, we revised the text, figure labels, and related descriptions (including Sect. 4.3 and associated figures) to consistently report glacier surface velocity in m d⁻¹ and removed the previous mixed usage of m yr⁻¹ / m/year.
Figure 7: Mean glacier surface velocity in the Kangri Karpo region (2015–2024).
Figure 8: Correlation between glacier area and velocity. Blue points denote glaciers in the study area; the red solid line is the fitted regression; axes are logarithmic.
Figure 9: (a), (c) Variations of glacier surface velocity and slope along the centerlines of the Yanong and Xirinongpu glaciers. Blue curve: velocity; red curve: slope. (b), (d) Scatterplots of velocity versus slope for all pixels within the Yanong and Xirinongpu glacier masks; red dashed line: linear fit of the slope-velocity relationship.
Specific comments
Title: “High resolution” could mean high spatial resolution and temporal resolution. As you indicated “monthly glacier surface velocity”, how about using “30-meter” to replace “highresolution”?
Thank you for this helpful suggestion. We agree that the term “high resolution” may be ambiguous because it can refer to either spatial or temporal resolution. To make the title more specific, we have revised it by replacing “high-resolution” with “30-meter”.
Revised title
30-meter monthly glacier surface velocity mapping in the Kangri Karpo region (2015–2024) using multi-source remote sensing data fusion
Line 17: This sentence confused me the fusion method is for whether high spatial resolution or high spatial and temporal resolution?
Thank you for this comment. We agree that the original wording was ambiguous and could be interpreted as referring to both spatial and temporal resolution. In the revised manuscript, we clarified this point by replacing “high-resolution” with “high-spatial-resolution” in the relevant sentence (Line 17) and throughout the manuscript where appropriate.
Line 25: It would be better to state as “with similar area and slope, south-facing glaciers are slightly faster than northern ones”.
Modified accordingly.
“…,within individual glaciers, velocity responds more strongly to slope; and, with similar area and slope, south-facing glaciers are slightly faster than north-facing ones.”
Line 96: Please add detail precipitation of this area to illustrate this area is “among the wettest sectors”. You also mentioned in Line 99 “high annual precipitation and humidity”, which may be repetition in meaning.
Thank you for this comment. We agree that the expression “among the Plateau’s wettest sectors” would be stronger if supported by explicit precipitation values. As the cited reference (Wu et al., 2021) provides a qualitative description rather than site-specific precipitation numbers for our study area, we avoided introducing unsupported quantitative values. Instead, we revised the sentence to a more cautious qualitative wording (“humid, monsoon-influenced sector”) and removed the potentially repetitive phrasing elsewhere in the manuscript. And by replacing the repeated climate description with a more specific statement on monsoon influence, seasonal hydroclimatic variability, and frequent cloud cover.
“Located at the eastern terminus of the Nyainqêntanglha Range on the southeastern Qinghai–Tibet Plateau, the Kangri Karpo region is a relatively wet sector of the Plateau (Wu et al., 2021).”
“Climatically, the area lies within the monsoonal temperate glacier region and is strongly influenced by the Indian monsoon (Yang et al., 2008), with pronounced seasonal hydroclimatic variability and frequent cloud cover.”
Figure 1: “KM” or “km”, but not “Km”.
Modified accordingly.
Figure 2: “Weight” not “Weigth”. Also, the use of blue and white in the figure is a bit misleading. UAV velocity image and Landsat velocity image are in white, which means they are processing and analysis steps as you indicate in the caption. What are the input datasets of them? Why the final outputs of the velocity results (2015-2024) are not in blue?
Modified accordingly.
Line 159: Please provide the detail acquisition time of the six UAV images, rather than listing
a time span of June to November in 2023.
Thank you for this suggestion. We have revised the manuscript to report the exact acquisition dates of the six UAV orthomosaics (8 June, 4 July, 10 August, 1 September, 3 October, and 20 November 2023), instead of only giving the June–November 2023 time span.
“In addition, we acquired UAV photogrammetry for reference and evaluation using a DJI Matrice 300 RTK equipped with an M6 Pro (M6P) metric mapping camera. Six orthomosaics were collected on 8 June, 4 July, 10 August, 1 September, 3 October, and 20 November 2023, forming five image pairs.”
Line 166: “32×32 pixels” not “32*32 pixels”. The results of “2 pixels for Landsat OLI” and “3 pixels for Sentinel-2 MSI” have the same spatial resolution. Why did you not set the sept size as 1 pixel for both Landsat and Sentinel-2, and resample the displacement result to 30m afterwards?
Thank you for this helpful comment. We have corrected the formatting from “32*32 pixels” to “32×32 pixels.”
关于步长设置，我们特意为Landsat OLI使用了2像素步长，为Sentinel-2 MSI使用了3像素步长，以便使位移输出的间距与目标约30米分辨率直接匹配。如果两个传感器都使用1像素步长，然后再重采样到30米，则会显著增加计算成本和处理时间。此外，与先生成密度更高的10/15米位移场，然后再重采样到30米相比，直接获取约30米分辨率的结果可以避免额外的重采样/插值步骤，并减少潜在的平滑效应，这更有利于后续的融合。
第 170 行：我建议在本段中更多地介绍时间序列的稳定性，包括 α 截尾均值滤波器的使用描述、选择 0.33 的阈值标准以及“*”的使用（这可能不是正文中的正式用法）。
感谢您的建议。我们在修改后的稿件中阐明了α = 0.33的设定依据。该参数是根据我们研究区域冰川速度时间序列的观测特征，通过经验方法选定的。在融雪季节（通常为6月至9月），光学遥感观测（Landsat和Sentinel-2）经常受到云雾污染（以及季节性积雪/云影效应）的影响，这会导致推导出的月平均速度出现异常高低的偏差。如果不加以抑制，这些异常值可能会严重干扰后续的融合结果。基于研究区域大部分年份的观测数据，年度月平均速度序列的中间部分提供了更为稳定的参考值，因此我们采用了α = 0.33的对称截尾方法（即截去下三分之一和上三分之一，并对中间约34%的数据取平均值），作为一种实用且稳健的设定。
对于Landsat和Sentinel-2影像，速度是通过COSI-Corr软件，利用频域互相关算法计算得到的。搜索窗口为32×32像素；Landsat OLI的步长为2像素，Sentinel-2 MSI的步长为3像素；相关阈值为0.95。东西向（EW）和南北向（NS）位移合并为总位移，然后除以成对的时间基线，得到30米分辨率的月度冰川表面速度。
为了提高可靠性，我们首先剔除了速度大于 3 md⁻¹ 的不匹配像素以及信噪比 (SNR) 小于 0.9 的低置信度像素。对于光学影像衍生的速度（Landsat 和 Sentinel-2），由于光学匹配更容易受到云/云影污染、季节性积雪、光照变化和地表纹理变化的影响，因此需要进行额外的稳定处理。这些因素可能导致某些月份出现异常大的或小的不匹配值。为此，我们应用了年内 α 截尾均值滤波器来稳定每个像素的月时间序列，并降低残余异常值的影响。具体而言，对于每个像素，我们对给定年份内的 12 个月速度值进行排序，并应用 α = 0.33 的对称截尾，即舍弃大约下 33% 和上 33% 的值，并将中间约 34% 的值取平均值，从而得到代表该年份的参考值。我们选择 α = 0.33 作为 12 个月年度序列的实用稳健设置，因为它既能抑制极端值，又能保留中心子集以实现稳定估计。然后，我们将月度值与该参考值逐像素进行比较；大于参考值 1.5 倍或小于参考值 0.5 倍的值被标记为异常值并移除。
第 311 行：“0.10，低于”而不是“0.10-低于”
已据此修改。
第 321 行：“product” 而不是 “produc”
已据此修改。
第 396 行：请使用连接号 (en dash)。“January–March” 而不是 “January-March”。
已据此修改。
-
AC1: 'Reply on RC1', Daoxun Gao, 23 Feb 2026
reply
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 126 | 72 | 14 | 212 | 30 | 9 | 11 |
- HTML: 126
- PDF: 72
- XML: 14
- Total: 212
- Supplement: 30
- BibTeX: 9
- EndNote: 11
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1