Plant phenology is influenced by various climatic factors such as temperature, precipitation, insolation, and humidity, etc. Among these factors, temperature and precipitation are proved to be the most important. However, the relative importance of these two factors is different among various phenophases and regions and is seldom discussed along environmental gradients. Based on normalized difference vegetation index (NDVI) data from the NDVI3g dataset and using the mid-point method, we extracted the start date of the growing season (SOG) and the end date of the growing season (EOG) in northern China during 1982–2012. To determine which climate factor was more influential on plant phenology, partial correlation analysis was applied to analyse the spatial difference between the response of SOG and EOG to temperature and precipitation. Finally, we calculated the temperature and precipitation sensitivities of the SOG and EOG. The results showed that: (1) SOG displayed an advancing trend in most regions, while EOG was delayed for all the vegetation types during 1982–2012. (2) SOG was mainly triggered by preseason temperature. The increase in temperature caused an overall advance in SOG. However, the relationship between SOG and precipitation varied among different vegetation types. Regarding EOG, precipitation had greater impacts than temperature in relatively arid environments, such as deserts, steppes and meadow biomes. (3) The response of vegetation phenology (both SOG and EOG) to temperature became stronger with increasing preseason precipitation across space. The response of EOG to precipitation became weaker from arid regions to relatively humid regions. These results provide a better understanding of the spatial pattern of the phenological response along the precipitation gradient and a reference for assessing impacts of future climate change on vegetation phenology, especially in transitional and fragile zones.

Different from other similar studies, in this paper, most of the data were excerpted from historical archives and documents, and were used to study the spatiotemporal dynamics of cultivated land resources in China during the last 300 years. It is clear that these data may help reveal land use change dynamics and its regional differences, but they may be problematic due to the constraints of the original data in the Ming Dynasty, Conversion of Land Area for the purpose of collecting land taxes, and the deliberate or inadvertent omission of cultivated land area during land surveys, therefore, such data were adjusted to our need. In processing the data, we made great efforts to analyze the historical context of their sources and reduced the possible errors. The results show that the cultivated land area increased most quickly in the early Qing Dynasty, and slowed down after the middle Qing Dynasty, and then was stable in the late Qing Dynasty until 1949, and has been decreasing since then. It is also found that the cultivated land use varied greatly in different regions. The east of the country was cultivated much more heavily than the west, but in some western provinces cultivated land area increased more quickly. It is considered that the driving factors of such cultivated land area change include the increase of the population, the political issues, and the impacts of wars. Natural environmental factors and the introduction of new crops might also have affected the cultivated land use change in the past 300 years.

Existing evidence of plant phenological change to temperature increase demonstrates that the phenological responsiveness is greater at warmer locations and in early-season plant species. Explanations of these findings are scarce and not settled. Some studies suggest considering phenology as one functional trait within a plant's life history strategy. In this study, we adapt an existing phenological model to derive a generalized sensitivity in space (SpaceSens) model for calculating temperature sensitivity of spring plant phenophases across species and locations. The SpaceSens model have three parameters, including the temperature at the onset date of phenophases (T_p), base temperature threshold (T_b) and the length of period (L) used to calculate the mean temperature when performing regression analysis between phenology and temperature. A case study on first leaf date of 20 plant species from eastern China shows that the change of T_p and T_b among different species accounts for interspecific difference in temperature sensitivity. Moreover, lower T_p at lower latitude is the main reason why spring phenological responsiveness is greater there. These results suggest that spring phenophases of more responsive, early-season plants (especially in low latitude) will probably continue to diverge from the other late-season plants with temperatures warming in the future.