Climate modeling of tree species distributions has demonstrated that future global climate change will have significant impacts on forest ecosystems [ 25– 26]. Discrepancies do occur between different climate modeling systems [ 27], but the approach nevertheless functions as an important research tool for assessing and predicting future changes in species distribution [ 28]. Despite the great strides in climate research made in recent years, most studies on plant species in China provide only approximate locations of change in habitat and rough descriptions of broad trends [ 29– 31]. In the present study, we performed a more detailed study on the suitable habitat of P. orientalis that will function as an important first step in developing strategies and policies for management and utilization of this important forest tree.

P. orientalis has a wide but fragmented distribution in China. It ranges across the southern sector of Inner Mongolia, Jilin and Liaoning, northern Shanxi, Gansu, Sichuan, Guizhou, Yunnan, Guangdong, and northern Guangxi [ 32]. It also occurs in Japan and North Korea. Based on large-scale environmental data, our modeling study projected a distribution of ca. 4.2×10 ⁶ km ² for the species under current climate conditions, in which the center areas are in mid-western Henan, the central and southern sectors of Shanxi and Shaanxi, eastern Gansu, northern Hubei, southwestern Jiangsu, central Shandong, southeastern Anhui, central Guangdong, and northern Yunnan. This model is in agreement with a previous study that located the most suitable areas of the species in typical sub-arid regions in western Henan, Shanxi, Shaanxi, and Gansu [ 33], but the potential range we identified was broader.

In our modeling exercise, temperature and wetness are among the most important factors affecting the distribution of P. orientalis. Previous studies have shown that seeds germination rate of P. orientalis decreases rapidly with decreasing of matric water potential [ 34]. Water availability is correlated with many environmental factors that influence the physiological and biochemical processes of plants, e.g. soil moisture is shown to be the main factor affecting the plant assimilation rate [ 35]. The number of branches and female flowers of P. orientalis are significantly associated with the average annual rainfall [ 36] while the male flower development are closely related to both temperature and precipitation [ 37]. On the other hand, the seedling emergence rate and death rate of P. orientalis are directly affected by temperature [ 36]. All these hydrothermal factors may have played main roles in shaping the ecological adaptation of P. orientalis, and have great impacts on the distribution of this conifer tree.

The intensity of UV radiation may also have a major role in shaping the distribution range of P. orientalis. UV-B reaching the ground has a great impact on the subaerial organs of plants [ 38]. It affects plant morphology by inhibiting stem elongation and leaf area expansion in many species subjected to elevated UV-B [ 39]. UV-B radiation decreases photosynthetic activities [ 40] and influences the functioning of plant protective mechanisms [ 41]. This form of radiation interacts with other environmental factors, such as water stress, increased CO ₂ concentration, and the availability of nutrients [ 42]. Levels of these factors modulate the effects of UV-B, which in turn may affect plant responses to other environmental parameters [ 43– 45]. Previous studies have shown that the light saturation point, net photosynthesis, photorespiration of P. orientalis are closely related to the component of light source [ 34]. However, little is currently known about the specific effects of UV-B radiation on the changes in the physiology biochemistry of P. orientalis, and we recommend quantitative studies in the future.

Mountain forest ecosystems in the Northern Hemisphere forest are likely to shift to higher elevations as the atmosphere warms [ 46– 48]. Consistently, our present study predicted upward shifts (Figs 1B and 3). It was predicted that under the low concentration greenhouse gas emissions scenario (RCP2.6), suitable habitat range will increase as global warming intensity proceeds. Under the higher concentrations of the high greenhouse gas emissions scenario (RCP8.5), we predicted a slight expansion in the next 40 years around 2050 followed by the slight contraction in the next 70 years around 2070. Overall, our prediction showed that the shift in distribution of suitable habitat areas to high latitudes and elevations would become gradually more significant, while low-latitude regions, such as Guangdong and Guangxi, would see a significant decrease in distribution. Our study emphasizes that forest surveillance should be planed for marginal populations both in low and high latitudes, so as to give a reasonable management advice in context of future climate change.

Although species are generally migrating to higher latitudes and elevations as climates warm [ 46, 49], the trends may differ between widely distributed species and those that have narrow ranges of occurrence. Plants with small distribution usually have a constrained ecological adaptability and are more susceptible to the impact of climate change. As an example, the suitable habitat area of the herb, Sinopodophyllum hexandrum, was predicted to fluctuate widely under future climate changes [ 29]. Plants with larger distribution may have broader adaptability and greater resistance to climate change. We found, under the future climate conditions, the distribution of P. orientalis increased slightly at first, but then began to decrease and fragment as the suitable areas shifted to higher latitudes and elevations during climate warming intensification; these trends are in line with those of previous studies [ 31, 50– 52]. For example, the suitable habitat area for Pinus massoniana will shrink [ 53– 54] and that of Pseudotaxus chienii will likely become fragmented [ 55] and reduced under climate change. Habitat fragmentation represents a major threat to tree populations by reducing local population size and gene flow from other populations, which in turn can decrease outcrossing rates, genetic variation [ 56– 59] and the adaptation to future climates. Little is known about the magnitude, direction or time scale of habitat fragmentation impacts on reproductive success and ultimately population viability of P. orientalis. In addition to shifts to higher latitudes and elevations, attentions should also be taken to fragmentation and shrinkage in habitats in evaluating the climate change impacts and developing a global change forest management strategy.

The combined effects of climatic changes and future land use transformations will increase unsuitable habitats more than one can predict [ 60]. Our analysis suggested that P. orientalis has a broad adaptability under the current and future climates, but much of the suitable area might disappear due to land use transformations and humans exploitation for economic purpose. Practically the entire suitable habitat of P. orientalis is transected by or within the vicinity of populated places, and much of the suitable area for P. orientalis has already been converted into agriculture or urbanized. Studies are still needed to qualify the anthropogenic impacts on the adaptation to future climates for this wide-spread conifer tree.

Seven sectors of high risk in the face of climate change have been identified in China: (i) northern semiarid, semihumid areas; (ii) the northwestern semiarid region; (iii) the North China Plain; (iv) Southern Hills; (v) the southwestern mountain region; (vi) the southwestern limestone mountain region; and (vii) the Qinghai-Tibet Plateau [ 61]. Areas that may become unsuitable for P. orientalis under the future climate scenario/year combination RCP2.6–2050 matched susceptible regions previously identified. Under combination RCP2.6–2070, regions losing the most suitable habitat area were the susceptible sectors previously identified as (iv) and (vi) above, and the South Hills area. Large losses of suitable habitat also occurred in sectors (iv) and (vi) under scenario/year combination RCP8.5–2050 and RCP8.5–2070. The predictions of distribution change in the present study were performed based on two extreme greenhouse emission scenarios (RCP2.6 and RCP8.5) from one General Circulation Model (GCM). The projections for future climate can be very different among GCMs. Nonetheless, we hope the extreme scenarios we molded will present a broadly forecast for the species´ range shift under future conditions. More modeling under a range of climate conditions are needed to reach a better resolution and a stronger recommendation for conservation operations.

Adaptation strategies to climate change, such as assisted migration, have gained increasing attention as we seek to help plants keep pace with changing climatic conditions [ 62]. Our modeling predicted an apparent shift of suitable habitat area for P. orientalis toward high latitudes and elevations, and also identified the low elevation regions in southern China facing high risks of habitat loss. We propose the following management strategy in response to future climate change: (1) implementation of seed collection and banking actions across the entire species range to provide a safety net in the face of natural and human uncertainty; (2) a focus on target populations likely to go extinct under climate change, especially those at low altitudes and latitudes; and (3) the use of seeds collected in the southern portion of the range to augment extant plantation in the northern portion, thereby introgressing potentially adaptive traits.

Based on Maxent modeling, we identified a broad habitat for P. orientalis under current climate conditions. The projected distribution matches well the extant range of the species across most of eastern part of China. Annual temperature, monthly minimum ultraviolet-B radiation (UV-B), wet-day frequency, and monthly maximum UV-B were important factors shaping the habitat availability for this species. Our model predicted that the shift of habitat to higher latitudes and elevations will become gradually more significant during climate warming intensification, while the species abundance in low-latitude regions will decrease. The projected spatial and temporal pattern of range shifts for P. orientalis will be a useful reference in developing forest management and conservation strategies for this ecologically important species.