the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 18 Aug 2025
Forest conversion reduces soil water retention in tropical rainforest by altering soil properties
Abstract. Extensive primary forests are being converted to secondary forests and plantation owing to human activities in recent decades, which has substantial effects on soil hydrological processes. However, the potential impact of forest conversion on soil water retention remains poorly understood. In this study, tropical primary forests (PF), secondary forests (SF) and rubber monocultures (RM) converted from tropical primary forests were selected on Hainan Island, to examine the variation in soil water retention across three forest types and their controlling factors. We found that the primary forests exhibited significantly greater water retention capacity than secondary forests and rubber monocultures. However, secondary forests showed higher water retention than rubber monocultures in shallow soils but lower in deep soils. Similarly, primary forests demonstrated significantly greater soil water storage capacity than secondary forests and rubber monocultures, but secondary forests and rubber monocultures had obvious seasonal variations, which showed that secondary forests had a higher water storage capacity than rubber monocultures in the rainy season, and display opposite pattern in the dry season. The saturated hydraulic conductivity in primary forests was higher than that in secondary forests and rubber monoculture. Furthermore, forest types influenced soil properties, with secondary forests and rubber monoculture showing higher bulk density but lower soil capillary porosity compared with primary forests. Among all factors, soil porosity emerged as the dominant controller of water retention, where total porosity and capillary porosity accounted for 31.49 % and 30.61 % of variation respectively, while soil bulk density contributed relatively less (12.46 %). These findings indicate that the conversion of tropical primary forests to secondary forests and rubber monocultures is detrimental to soil water retention and storage. Our results can provide scientific insights for forest development and management in the tropical rainforest.
- Preprint
(1367 KB) - Metadata XML
- BibTeX
- EndNote
Status: open (until 18 Jan 2026)
-
RC1: 'Comment on egusphere-2025-3772', Anonymous Referee #1, 10 Oct 2025
reply
-
AC1: 'Reply on RC1', Licong Dai, 24 Oct 2025
reply
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2025/egusphere-2025-3772/egusphere-2025-3772-AC1-supplement.pdf
-
AC1: 'Reply on RC1', Licong Dai, 24 Oct 2025
reply
-
RC2: 'Comment on egusphere-2025-3772', Anonymous Referee #2, 04 Jan 2026
reply
General comments
The manuscript titled “Forest conversion reduces soil water retention in tropical rainforest by altering soil properties” essentially extends the work previously published by the same authors in 2024 (Chen et al., 2024; 10.1093/jpe/rtae021). In their earlier study, the authors compared soil physical and hydraulic properties between two sites, secondary forest and rubber plantation, using soil samples collected during field surveys conducted in 2022 in Baoting County, Hainan Island, China. In the present manuscript, a third site, including primary forest, is added to the comparison. Although the objective of comparing soil properties across different land uses is undoubtedly of interest, the paper does not convincingly extend the discussion initiated in the previous work. The manuscript would benefit from a more critical and explicit analysis of how the inclusion of the primary forest site advances understanding beyond the initial study.
In addition, comparing Fig. 1 of the present paper with Fig. 1 of Chen et al. (2024), it appears that the secondary forest sites are located in different positions. However, the sites are reported to share exactly the same stand characteristics (as shown in Table 1 of both studies). It is unclear whether this similarity reflects the selection of homogeneous areas or if the sites are actually the same. Since neither paper reports the exact site coordinates, and the present manuscript does not explicitly clarify whether the sites coincide, readers may be confused when comparing the two studies by the same authors.
Because of these and other shortcomings noted in the following comments, my recommendation is that the manuscript cannot be accepted in its current form and should be thoroughly revised before resubmission.
Specific comments
LL 93-94. The authors state that, to ensure comparability among the study sites, all selected plots shared similar biophysical conditions, including altitude, slope, and aspect. However, inspection of Fig. 1 suggests that this assumption may not be fully met, as at least the altitude of the tree plots appears to differ substantially among sites. Such divergence in elevation could influence key environmental drivers (e.g., temperature, precipitation patterns) and therefore potentially confound the interpretation of the reported results. In the Study Site section, the authors report basic climatic information, namely mean annual rainfall and temperature, for Baoting County as a whole. However, given the evident differences in altitude among the study sites, readers may reasonably question whether local climatic conditions are truly comparable. Elevation gradients can induce significant variations in temperature and precipitation regimes. Are site-specific or nearby meteorological data available to substantiate the authors’ assertion of similar climate conditions across the study sites (LL 96)? The authors should clarify the magnitude of these differences and justify why they do not affect site comparability, or explicitly account for altitude as a source of variability in the analysis.
LL 100-103. The authors do not explicitly report the number of collected soil cores. Based on the description of the field surveys, it can be inferred that a total of 2,592 soil cores were collected (3 sites × 3 plots × 4 subplots × 6 depths × 12 months), representing an impressive sample size. Does this estimate correspond to the actual sample size? The authors are encouraged to explicitly report the total number of samples collected, as well as the number of disturbed samples used for laboratory physical and chemical analyses. In addition, it should be clarified whether these disturbed samples were also collected at multiple depths.
LL 115. Did the authors adopt a specific protocol to minimize or prevent air entrapment in the soil samples? According to the referenced paper (Chen et al., 2024), it appears that the collecting rings containing the samples were simply immersed in a bucket with water up to the upper edge. This procedure may not be sufficient to avoid air entrapment within the soil, which could affect subsequent measurements. The authors should clarify whether any additional measures were taken to ensure complete saturation and avoid air entrapment.
LL 119. The authors refer to their own recently published paper to support the applied standard laboratory analyses. In this context, the use of self-citation should be avoided. Instead, the authors are encouraged to cite more established and widely recognized references that describe standard analytical procedures. In addition, the adopted saturate-and-drain procedure raises some concerns regarding the determination of field capacity. The exact drainage time required to reach equilibrium is strongly soil-dependent: coarse-textured soils may equilibrate within approximately 24 h, whereas fine-textured or clayey soils often require substantially longer drainage periods. Therefore, the use of a fixed drainage time may lead to inconsistent or biased estimates of field capacity across different soil types. Moreover, this approach does not strictly conform to the classical definition of field capacity, which is generally defined at a specific matric potential (typically around −33 kPa). The authors should justify the selected drainage time in relation to soil texture or consider using matric-potential-based methods to ensure a more robust and comparable estimation of field capacity.
Citation: https://doi.org/10.5194/egusphere-2025-3772-RC2
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 1,040 | 59 | 30 | 1,129 | 40 | 40 |
- HTML: 1,040
- PDF: 59
- XML: 30
- Total: 1,129
- BibTeX: 40
- EndNote: 40
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
This manuscript presents measurements of soil water holding capacity in primary forest, secondary forest, and rubber plantation in the humid tropical island of Hainan, China. The main findings are that primary forest has significantly higher water holding capacity than the disturbed landscapes, and that these differences are associated with higher macroporosity. The manuscript includes interesting time series measurements of measurements of water content and soil hydraulic properties during the wet season and through the transition into the dry season. The hydraulic differences between secondary forest and rubber plantation are also somewhat discussed. The main strength of this manuscript is an interesting and quite complete dataset of soil hydraulic and chemical properties across the three sites, and with some time series component.
This manuscript has significant shortcomings in both the scientific novelty and the analysis. Beginning with scientific novelty, it has been well-established for several decades that forest conversion leads to soil compaction and reduced hydraulic function, e.g., Bruijnzeel, L. A. 1990. Hydrology of Moist Tropical Forests and Effects of Conversion: A State of Knowledge Review; Bonell, M., and L. A. Bruijnzeel. 2005. Forests, Water and People in the Humid Tropics. Both these reviews and the works cited within provide in-depth exploration of the same topics presented in this current manuscript. The main findings of this paper show quite extensively that several different soil water retention properties are related to several different soil porosity parameters - Figures 5, 7, and 8. More recently, studies cited within this manuscript also found the same results - Wen et al., 2017 and 2019.
Perhaps just as importantly, the analysis of the data has several shortcomings that hinder interpretation and comparability with the literature. First, three sites are used in total in a space for time approach - however, the land use history of the three sites is not mentioned in the manuscript. Moreover, the three sites appear to have considerable differences in their climate and geological setting, although the manuscript claims otherwise. Figure 1 indicates that though the sites are within a few miles of each other, each site has significantly slope and topographic setting. Particularly, the primary forest site appears to be up to 1000 meters higher altitude than the rubber monoculture site in the lowlands at roughly sea level. Accordingly, soil textural and mineralogical composition at the site appears significantly different (Figures 2 and 3). The manuscript indicates that the changes in soil texture are caused by the land use differences (line 146-147), but this link is more likely confounding, not causal. Furthermore, as a reviewer I am speculating that the precipitation would differ between the three sites, especially due to the elevation gradient. This is not acknowledged or discussed even though water storage is presented as a primary finding. In general, the confounding differences between the sites need to be addressed as a major limitation in the interpretation of results and especially causality.
I also want to briefly address line 43-44: “However, in recent decades, economic development and slash-and-burn cultivation by ethnic minorities have led to extensive degradation of primary forests”. It is wholly inappropriate and scientifically irrelevant to comment on ethnic minorities as a cause of deforestation.
This project’s strength lies in the data collection performed over time. Figure 6A is genuinely interesting, particularly the wide difference in soil water storage during the dry season. Linking this behavior to observed soil traits (and likely precipitation differences) would provide useful new insights into ecosystem hydrological function. This may be a path to publication in the future, but the manuscript in its current form is not suitable for publication in EGU SOIL.
Specific comments: