**5. Conclusions**

In this study, we surveyed the vegetation parameters, topographic variables, and soil properties in the dry and rainy seasons from 2017 to 2018 in the arid-hot valley in southwestern China, and analyzed the relationships among them by using Pearson's correlation coefficient and redundancy analysis. Our results sugges<sup>t</sup> that soil moisture was relatively adequate during the rainy season, but that only *D. viscosa* survived on sunny slopes at lower elevations ( ≤1520 m) near the end of the dry season because of the low average soil moisture. The lack of large differences in soil fertility among the elevations sugges<sup>t</sup> that soil fertility would have a relatively small influence on vegetation restoration. The biomass of *D. viscosa* on the shade slopes initially increased with increasing elevation, then decreased again as conditions became unsuitable for the species, but decreased steadily with increasing elevation on the sunny slopes. By contrast, the biomass of *P. yunnanensis* at low elevations (≤1640 m) was higher on the shade slopes than on the sunny slopes, but it was less at high elevation (1940 m). RDA analysis showed that topographic conditions explained vegetation variation better than soil conditions for both species, but that changes in soil moisture along the elevation gradient at the end of the dry season strongly affected vegetation restoration. Our study provides important scientific support for planning ecological restoration in our study area.

**Supplementary Materials:** The following are available online at http://www.mdpi.com/2071-1050/10/12/ 4774/s1, Figure S1: Distribution of temperatures and precipitation for Yongsheng County meteorological station. Variable names: Jan, January; Feb, February; Mar, March; Apr, April; Jun, June; Jul, July; Aug, August; Sep, September; Oct, October; Nov, November; Dec, December, Table S1: Topography conditions and soil nutrient elements of quadrats in study area, Table S2: Correlation analysis (Pearson's *r*) for the relationships among the vegetation and environmental factors for *Dodonaea viscosa* and *Pinus yunnanensis*. Significance: \*, *P* < 0.05; \*\*, *P* < 0.01. Bio: Biomass; Den: density; Bra: branches; Cro: crown; Hei: height; Cov: coverage; Asp: aspect slope; Slo: slope; Ele: elevation; DBH: diameter at breast height; DQR: distance from quadrates to river; TWI: topographic wetness index; SM: soil moisture; OM: organic matter; TN: total nitrogen; TK: total potassium; AN: hydrolyzed nitrogen; AP: available phosphorus; SF: soil fertility.

**Author Contributions:** Conceptualization, W.Y.; Data curation, W.Y.; Formal analysis, W.Y.; Funding acquisition, Y.C.; Investigation, J.P., Y.Y. and X.L.; Methodology, J.P. and X.L.; Project administration, Y.C.; Resources, Y.C.; Software, J.P. and X.L.; Supervision, W.Y.; Validation, Y.Y.; Writing—original draft, J.P.; Writing—review & editing, W.Y. and Y.Y.

**Funding:** National Key Research and Development Plan of China (No. 2016YFC0502209 and 2017YFC0404505), and the National Natural Science Foundation of China (No. 51579012).

**Acknowledgments:** We thank the National Key Research and Development Plan of China (No. 2016YFC0502209 and 2017YFC0404505), the Beijing Nova Program (No. Z171100001117080), and the National Natural Science Foundation of China (No. 51579012) for their financial support. We also thank Geoffrey Hart for providing language help during the writing of this paper.

**Conflicts of Interest:** The authors declare no conflicts of interest.
