Estimating suitable hydrothermal conditions for vegetation growth for land use cover across China based on maximum-probability-density monthly NDVI

• Published in: Remote sensing applications 2023-04, Vol.30, p.100958, Article 100958
• Main Authors: Shen, Chenhua, Ma, Riran
• Format: Article
• Language: English
• Subjects: Land cover types; Maximum-probability-density; Simulation of NDVI changes; Suitable hydrothermal condition; Vegetation growing period
• ISSN: 2352-9385; 2352-9385
• DOI: 10.1016/j.rsase.2023.100958

To provide a theoretical basis and key parameters for ecological process modeling, the maximum-probability-density (MPD) atmospheric temperature and precipitation on vegetation growth for various land cover types throughout China were estimated using a non-parametric statistical technique, based on MPD monthly Normalized Difference Vegetation Index (NDVI) during 2000–2021. The results revealed that more than half of all MPD monthly temperatures deviated from the average monthly air temperature. The spatial distribution of MPD monthly temperature closely resembled that of traditional temperature zones. The deviation from the average temperature was caused by the physiological rhythm of vegetation growth. Less than half of all MPD precipitations deviated from the average monthly precipitation. The deviation was caused by the ability of vegetation to absorb water. The suitable temperatures and precipitations on vegetation growth for vegetation cover types, temperature zones, and dry-wet zones were unique and varied widely. The same vegetation in different temperature zones and dry-wet zones had different MPD temperatures and precipitation. In the scenario of Representative Concentration Pathway (RCP) 2.6, global warming in the future would have negative impacts on vegetation growth in arid and semi-arid regions but have positive effects on vegetation growth in the Qinghai-Tibet Plateau. The rise in precipitation could partially offset the negative effects of rising temperatures. The parameters in the Gaussian kernel density estimation method and the traits of the original dataset (spatial resolution, variance, and length) were the primary influencing factors on the uncertainty of spatial distributions. Our results highlight the potential of MPD temperatures and precipitations in ecological process modeling, particularly in the simulation of NDVI under the influence of air temperature and precipitation changes.