4.2 Spatial transformation characteristics of T. chinenseunder climate change scenarios
In this study, Biomod2 was used to predict the distribution of T. chinense in Asia, and the results were not completely consistent with the regional niche results predicted by Tang et al. and Gao et al. using the MAXENT model (P. Gao et al., 2023; Tang et al., 2021). It is speculated that there are four reasons for this result. The first point is the choice of modeling scope. In the existing studies, the methods of constructing species distribution models are divided into complete and regional niche models. Complete niche models are more reliable models for describing the climatic niche of species because they can calibrate the presence and capture a wider range of occupied environmental conditions (Pearson et al., 2004). However, if unoccupied but environmentally suitable areas for the species are considered for model training, then the capacity to predict the species’ potential distribution will be reduced (Acevedo et al., 2012). Barve et al. argued that the appropriate geographical background for model training, validation, and comparison should comprise the set of localities that a species has over its history (Barve et al., 2011). In this study, the geographical distribution of T. chinense was reviewed, and the region with the highest accessibility (Asia) was selected as the study area. In addition, the use of ensemble species distribution model can reduce the instability of a single model (Goldsmit et al., 2020; Araújo & New, 2007), and can more accurately predict the range of suitable areas of T. chinense . Secondly, there is a serious bias in the geographical distribution of the occurrence data used by Tang and Gao et al. in constructing the model (P. Gao et al., 2023; Tang et al., 2021). Through our field investigation, we found T. chinense populations also distribute in the high-latitude areas in the northern of China. However, in the studies of Tang et al and Gao et al., there are few or even no occurrence data of T. chinense in high latitude areas, relatively concentrated data in low latitude in the southern of China, which is biased from the real occurrence data of T. chinense . This biased regional data cannot objectively identify the niche of species (El‐Gabbas & Dormann, 2018). In addition to the above factors, we speculate that the choice of factors is also one of the reasons for the different results. As a semi-parasitic plant, vegetation (host) is the premise for the good growth of T. chinense , so biological factors play a certain role in the distribution modeling of the species. At present, the inclusion of interspecific interactions is considered to be the main challenge for species distribution modeling (Y. Wang et al., 2022). Considering that there are many host species of T. chinense , mainly Asteraceae, Fabaceae, and Poaceae, and the host distribution is extensive (Guo & Luo, 2011), we included NDVI data reflecting vegetation growth in the species distribution model. Although NDVI data can not describe the specific conditions of vegetation classification groups in the land acquisition table, we believe that the inclusion of NDVI data may make the modeling results closer to its real niche for T. chinense .
Compared with the current, the suitable areas of T. chinense show an increasing trend under the combination of future climate scenarios. Spatially, consistently with the migration direction of most plant species (B. Chen, Zou, et al., 2022; W. Li et al., 2019; Yin et al., 2022; L. Zhang et al., 2022), the overall distribution areas gradually move to the northwest of the current suitable areas (high latitude and high altitude areas). In the future, the suitable areas of T. chinense in China will mainly expand to Tibet. The expansion area is currently located in the semi-arid area, and its soil moisture content can meet the needs of the growth of T. chinense . However, due to the high altitude, its temperature is lower than other semi-arid areas suitable for the growth of T. chinense . Therefore, this area is not suitable for the growth of T. chinense in the current climate. In the future, the plateau temperate zone in the southern of China will expand northward, gradually replacing the plateau sub-cold zone in western China, and gradually increasing the temperature in this region (S. Wu et al., 2010). Therefore, the climate conditions in this area may be similar to those in the current suitable areas of T. chinense in the future, which may be one of the reasons for the expansion of the suitable areas in this area. In the future, the disappearing areas of suitable areas of T. chinense are mainly located in the southern of China, which will change from the subtropical zone to the tropical zone with climate warming in the future (S. Wu et al., 2010). At the same time, the increasing temperature will reduce the moisture of the surface soil, increase evaporation, reduce the Climate Moisture Index, and is not conducive to the absorption of soil fertility (Michaelian et al., 2011; Peng et al., 2011). This will make the areas no longer suitable for the growth of T. chinense .