Chir pine (Pinus roxburghii, Sarg.) forests are dominant in the Indian Himalayan region and act as a huge carbon (C) sink. However, measuring the C sink in soil is complex and time-intensive, and therefore the present study attempts to estimate the soil organic carbon (SOC) through a remote sensing (RS) approach. We estimated SOC stock of chir pine forests along an altitudinal gradient at three soil depths (0–30, 30–60 and 60–100 cm) in the Garhwal Himalaya, Uttarakhand. Fourteen forest stands at four altitudes, viz., <1000 m above sea level (m asl), 1001–1400 m asl, 1401–1800 m asl and >1801 m asl were surveyed and served for data collection. A model for predicting SOC was developed through stepwise regression analysis based on vegetation information and altitude as independent variables with the field data on SOC. For vegetation information, we used the normalized difference vegetation index (NDVI) measured through remote sensing (RS). The mean SOC stock up to 100 cm depth was increased with increasing altitude and were in the order of 69.66 ± 19.86, 85.27 ± 17.53, 95.68 ± 7.90 and 148.41 ± 71.39 million g ha⁻¹ (million gram per hectare) for <1000, 1001–1400, 1401–1800 and >1801 m_asl, respectively. The result showed that NDVI was a good predictor for SOC estimation. The model predicted SOC stock between 57 and 152 million g ha⁻¹ with a mean of 93 million g ha⁻¹, which was close to the SOCs from field inventory. Therefore, RS could be used to precisely map the SOC stock in the chir pine forests of the Himalayas through NDVI and provide information to policymakers for forest management.

中文翻译：

---

印度 Garhwal Himalaya 松树林海拔梯度的土壤有机碳估算：遥感方法的实地清单

杉木 ( Pinus roxburghii ), Sarg.) 森林在印度喜马拉雅地区占主导地位，并充当巨大的碳 (C) 汇。然而，测量土壤中的碳汇是复杂且耗时的，因此本研究试图通过遥感（RS）方法估计土壤有机碳（SOC）。我们估计了北阿坎德邦加瓦尔喜马拉雅山三个土壤深度（0-30、30-60 和 60-100 厘米）沿海拔梯度的红松林 SOC 存量。调查了四个高度的十四个森林，即海拔<1000 m (m asl)、1001-1400 m、1401-1800 m和>1801 m asl，并用于数据收集。以植被信息和海拔高度为自变量，结合土壤有机碳的实地数据，通过逐步回归分析建立了土壤有机碳预测模型。有关植被信息，我们使用通过遥感（RS）测量的归一化植被指数（NDVI）。直至 100 厘米深度的平均 SOC 存量随着海拔的升高而增加，依次为 69.66 ± 19.86、85.27 ± 17.53、95.68 ± 7.90 和 148.41 ± 7139 万克公顷^-1（百万克每公顷）分别表示 <1000、1001–1400、1401–1800 和 >1801 m_asl。结果表明，NDVI 是 SOC 估计的良好预测指标。该模型预测 SOC 库存在 57 到 1.52 亿 g ha ^-1之间，平均值为 9300 万 g ha ^-1，这接近于来自实地库存的 SOC。因此，RS 可用于通过 NDVI 精确绘制喜马拉雅松树林中的 SOC 库，并为森林管理决策者提供信息。