Institute of Mountain Hazards and Environment, Chinese Academy of Sciences, Chengdu 610299, China; University of Chinese Academy of Sciences, Beijing 100049, China.
Institute of Mountain Hazards and Environment, Chinese Academy of Sciences, Chengdu 610299, China.
Sci Total Environ. 2024 Dec 1;954:176666. doi: 10.1016/j.scitotenv.2024.176666. Epub 2024 Oct 8.
Satellite-derived products and field measurements verify that Tibetan Plateau (TP) has been experiencing continuous vegetation greening and productivity increase; however, it remains unclear how this greening translates into productivity and how long-term productivity variations depend on greenness across the TP. Moreover, ignoring the accuracy evaluation of satellite-derived greenness and productivity products may mislead the understanding of TP vegetation changes. Thus, we initially assessed the accuracy of three widely used leaf area index (LAI, proxy of greenness) products (i.e., MODIS, GLASS and GEOV2 LAI) and three gross primary productivity (GPP, proxy of productivity) products (i.e., MODIS, GLASS and PML-V2 GPP) to selected robust products to represent greenness and productivity respectively. Then, we explored the dependence of spatiotemporal GPP dynamics on greenness variations during 2000-2020. Results indicated that PML-V2 GPP and MODIS LAI were more robust and reliable than other satellite-derived products when compared to the reference values. They revealed a prevailing increase in GPP over the past two decades, with a regional average of 71 % higher than that of LAI. Notably, the area proportion of significant productivity enhancement was 31.6 % higher than that of significant greening. About 24.7 % of the TP displayed significantly inconsistent trends. The dependency of GPP on LAI gradually decreased with the increasing water availability, the complexity of vegetation structures, and dense canopy community. By calculating leaf photosynthetic capacity, we found that this indicator greatly regulated the velocity discrepancy between GPP and LAI, and the contribution of only greening to productivity is limited, only occupying 11.9 % of the TP, which was helpful in understanding the inter-annual changes of vegetation dynamics under varying environment conditions. We therefore reveal an unexpected rapid increase in productivity than greening during 2000-2020 on the TP, as well as highlight the caution of only using satellite-derived greenness indicators for assessing long-term changes in vegetation productivity dynamics, especially over mesic ecosystems with complex vegetation structures and dense canopies of TP.
卫星衍生产品和实地测量验证了青藏高原（TP）一直在经历持续的植被绿化和生产力提高；然而，目前尚不清楚这种绿化如何转化为生产力，以及长期生产力变化在多大程度上依赖于 TP 的绿化程度。此外，忽略卫星衍生绿化和生产力产品的准确性评估可能会导致对 TP 植被变化的误解。因此，我们首先评估了三种广泛使用的叶面积指数（LAI，绿化的代表）产品（即 MODIS、GLASS 和 GEOV2 LAI）和三种总初级生产力（GPP，生产力的代表）产品（即 MODIS、GLASS 和 PML-V2 GPP）对选定的可靠产品的准确性，以分别代表绿化和生产力。然后，我们探索了 2000-2020 年间时空 GPP 动态对绿化变化的依赖关系。结果表明，与其他卫星衍生产品相比，PML-V2 GPP 和 MODIS LAI 更稳健可靠。它们揭示了过去二十年中 GPP 的普遍增加，区域平均值比 LAI 高出 71%。值得注意的是，生产力显著增强的面积比例比绿化显著增加的面积比例高 31.6%。大约 24.7%的 TP 显示出明显不一致的趋势。随着水分可用性的增加、植被结构的复杂性和茂密的冠层群落，GPP 对 LAI 的依赖性逐渐降低。通过计算叶片光合能力，我们发现该指标极大地调节了 GPP 和 LAI 之间的速度差异，而绿化对生产力的贡献是有限的，仅占 TP 的 11.9%，这有助于理解在不断变化的环境条件下植被动态的年际变化。因此，我们揭示了在 2000-2020 年期间，TP 上生产力的快速增长超过了绿化，同时强调了仅使用卫星衍生的绿化指标评估植被生产力动态的长期变化时需要谨慎，特别是在具有复杂植被结构和茂密冠层的 TP 中湿润生态系统。
