Detailed analyses are presented of cropping intensity (CI) and crop-seasonality parameters, such as the start and end of a season, the length of a season, and the seasonal amplitude for the period 1999–2010 across Mainland Asia. The analyses used fitted Normalized Difference Vegetation Index (NDVI) time-series data derived from SPOT VEGETATION 10day synthesis (VGT S10) instruments. Savitzky-Golay noise filtering was applied to the NDVI time series, and the results of an automated extraction procedure were compared with the results of other research analyses. The results indicated that: 1) the projected spatial distribution of CI agreed with other analyses; 2) detected double-cropping regions generally extend over large areas equipped for irrigation; 3) the developed extracting algorithm was capable of estimating CI within Mainland Asia; 4) measured spatial variation at the start of a season adequately represented the crop calendar for each grid and region; 5) low standard deviations (SD) for the start of seasons and seasonal amplitude of NDVI across the period 1999–2010 were closely related to the presence of irrigated cropland, and SD values were relatively high for arid, rainfed zones; 6) there were large elements of uncertainty in estimations of cropping-seasonality parameters caused by (i) cloud contamination of images, (ii) the single setting of the smoothing filter and determinations of seasons’ starts/ends, and (iii) spatial and temporal resolution of satellite data. Detailed analyses of crop seasonality will contribute to progress across a range of agricultural concerns, including local and large-scale food security and the management of complex agricultural systems.

中文翻译：

---

多时相 SPOT 植被数据提取的亚洲作物强度和季节性参数

详细分析了种植强度 (CI) 和作物季节性参数，例如季节的开始和结束、季节​​的长度以及 1999-2010 年亚洲大陆的季节幅度。分析使用从 SPOT VEGETATION 10 天合成 (VGT S10) 仪器得出的拟合归一化​​植被指数 (NDVI) 时间序列数据。将 Savitzky-Golay 噪声过滤应用于 NDVI 时间序列，并将自动提取程序的结果与其他研究分析的结果进行比较。结果表明：1）CI的投影空间分布与其他分析一致；2) 检测到的双熟区一般都覆盖有灌溉设施的大片区域；3）开发的提取算法能够估计亚洲大陆的CI；4) 在一个季节开始时测量的空间变化充分代表了每个网格和区域的作物日历；5) 1999-2010 年期间 NDVI 季节开始的低标准差 (SD) 和季节幅度与灌溉农田的存在密切相关，干旱、雨育区的 SD 值相对较高；6）由于（i）图像的云污染，（ii）平滑滤波器的单一设置和季节开始/结束的确定，以及（iii）空间和卫星数据的时间分辨率。对作物季节性的详细分析将有助于解决一系列农业问题，包括地方和大规模粮食安全以及复杂农业系统的管理。