Using leaf area index (LAI) data from 1981 to 2014 in the tropical moist forest eco-zone of South America, we extracted start (SOS) and end (EOS) dates of the active growing season in forest and savanna at each pixel. Then, we detected spatiotemporal characteristics of SOS and EOS in the two vegetation types. Moreover, we analyzed relationships between interannual variations of SOS/EOS and climatic factors, and simulated SOS/EOS time series based on preceding mean air temperature and accumulated rainfall. Results show that mean SOS and EOS ranged from 260 to 330 day of year (DOY) and from 150 to 260 DOY across the study region, respectively. From 1981 to 2014, SOS advancement is more extensive than SOS delay, while EOS advancement and delay are similarly extensive. For most pixels of forest and savanna in tropical moist forest eco-zone, preceding rainfall correlates predominantly negatively with SOS but positively with EOS, while the relationship between preceding temperature and phenophases is location-specific. In addition, preceding rainfall is more extensive than preceding temperature in simulating SOS, while both preceding rainfall and temperature play an important role for simulating EOS. This study highlights the reliability of using LAI data for long-term phenological analysis in the tropical moist forest eco-zone.

中文翻译：

---

检查南美洲热带湿润森林生态区的地表物候

利用南美洲热带湿润森林生态区 1981 年至 2014 年的叶面积指数 (LAI) 数据，我们提取了每个像素的森林和稀树草原活跃生长季节的开始 (SOS) 和结束 (EOS) 日期。然后，我们检测了两种植被类型中 SOS 和 EOS 的时空特征。此外，我们分析了 SOS/EOS 年际变化与气候因素之间的关系，并基于前期平均气温和累积降雨量模拟了 SOS/EOS 时间序列。结果表明，整个研究区域的平均 SOS 和 EOS 范围分别为一年中的 260 至 330 天 (DOY) 和 150 至 260 DOY。从 1981 年到 2014 年，SOS 推进比 SOS 延迟更广泛，而 EOS 推进和延迟同样广泛。对于热带湿润森林生态区的森林和稀树草原的大部分像素，前期降雨主要与 SOS 呈负相关，但与 EOS 呈正相关，而前期温度与物候期之间的关系因地点而异。此外，在模拟SOS时，前期降雨比前期温度更广泛，而前期降雨和温度对模拟EOS都起着重要作用。本研究强调了使用 LAI 数据在热带潮湿森林生态区进行长期物候分析的可靠性。