Impact of evapotranspiration on ecosystem services values in rubber (Hevea brasiliensis L.) plantations: insights into climate change and rubber plantation expansion in Southwestern China

Introduction: The rapid expansion of rubber (Hevea brasiliensis) plantations in Southwestern China has transformed forest-dominated landscapes. However, the hydrological consequences of this land-use change have not been comprehensively quantified.

Methods: This study reconstructed the long-term actual evapotranspiration (ET_{a_reconstructed}) of rubber plantations using the "Kc-ET0" method, which integrates meteorological data and actual crop coefficients (Kc,actual). The coefficient of variation (CV) and the Random Forest (RF) algorithm were employed to identify the key drivers and analyze the spatiotemporal patterns of the reconstructed ET_{a_reconstructed}. Furthermore, we quantitatively evaluated the influence of ET_{a_reconstructed} on major ecosystem service values (ESVs).

Results: Our results showed that rubber plantations expanded at a rate of 12.49 × 10⁴ ha per decade between 2000 and 2022. The annual average ETa_reconstructed was 952 mm, showing an increasing trend of 12 mm per decade since 1970. The ETa_reconstructed exhibited stable interannual variability (CV range: 0.04–0.11) and was significantly positively correlated with maximum temperature, sunshine duration (SSD), and rubber plantation area, but negatively correlated with relative humidity. SSD (28.5%–38.4%) and rubber plantation area (11.4%–20.3%) were identified as the crucial drivers. The optimal ESVs for rubber plantations were achieved at a multi-year average SSD of 4.17 hours, where the reconstructed ETa ranged between 2.0 and 3.0 mm/day, and carbon sequestration and oxygen release reached their peak values.

Discussion: In conclusion, the ETa_reconstructed of rubber plantations, primarily driven by increased sunshine duration, significantly affects their ecosystem service values in Southwestern China. The identified optimal ETa_reconstructed range provides a scientific basis for precise water management in tropical rubber plantations in the context of climate change.

1 Introduction

Crop evapotranspiration is a critical component of land–atmosphere interactions that is closely coupled with energy and water cycles (Cutting et al., 2024). Land use/cover change (LUCC) is a major driver of global environmental change that impacts global carbon cycling and biodiversity. In tropical regions, the conversion of natural forest to commercial plantations significantly alters regional hydroclimates (Wang and Zhang, 2025; Lam et al., 2022). These transformations are most pronounced in tropical areas, such as Southeast Asia, the Amazon, and Africa (Chen et al., 2024; Sarathchandra et al., 2021). LUCC influences evapotranspiration by altering the surface energy balance, making it a direct indicator of hydrological cycle disruptions. Rubber (Hevea brasiliensis L.) plantations, as typical human-managed ecosystems, provide ecosystem service values (ESVs) such as water conservation, carbon storage, soil retention, and climate regulation (Qi et al., 2023). Evapotranspiration, a key indicator of vegetation water consumption, governs the hydrological regulation function of rubber plantations and constrains their overall ESVs. Excessively high evapotranspiration can lead to regional water overconsumption and reduced water conservation capacity, while excessively low evapotranspiration may limit plantation productivity and impair carbon sequestration function (Li et al., 2024). Therefore, a significant trade-off exists between evapotranspiration and the ESVs of rubber plantations.

Driven by rising market prices, rubber tree cultivation has expanded into non-traditional zones at higher latitudes (22°–23°N) and altitudes (up to 1,400 m), replacing extensive natural forests and farmland (Ling et al., 2022a). As a result of this expansion, ecological service functions have declined to 21.5%–63.6% of those in primary forests (Singh et al., 2021). In Southwest China, rubber plantations now cover over 2.2 million ha, which has reduced tropical rainforest cover by approximately 50% and intensified regional aridification and water conservation losses (Thellmann et al., 2019). Rubber plantations have been called “green deserts” to highlight the severe ecological consequences of their soil erosion rates (Ziegler et al., 2009; Liu et al., 2015). The high evapotranspiration of rubber plantations profoundly disrupts regional water balances through substantial water depletion and reduced soil water storage (Tan et al., 2011; Giambelluca et al., 2016). Rubber-dominated landscapes exhibit significantly greater water loss than other types of land cover (Lin et al., 2022; Zeng et al., 2022), which results in diminished streamflow, more frequent zero-flow days, and seasonal water shortages (Xue et al., 2024; Kumagai et al., 2015). Consequently, optimizing evapotranspiration management for climate change is critical to sustain the ecological services of rubber plantation ecosystems.

More attention has been paid to the eco-hydrological consequences of rubber plantation expansion. Giambelluca et al. (2016) reported high evapotranspiration rates in rubber plantations across Thailand and Cambodia and highlighted concerns over water consumption patterns and the potential risks to regional water security with continued expansion. The study also underscored the importance of balancing ecological considerations with the socioeconomic benefits of plantations. Comprehensive assessments of the ecosystem services inherent to rubber plantations are scarce. Kiyono et al. (2014) proposed a potential role for rubber plantations in climate change mitigation, and several investigations have addressed carbon sequestration, oxygen release, and timber utilization. Qi et al. (2023) extended this work by evaluating the integrated socioeconomic and environmental impacts of carbon, oxygen, and biodiversity services in Chinese rubber plantations. The evapotranspiration of rubber plantations has a crucial impact on the value of ecological services, but current studies are not clear.

To comprehensively assess the impacts of evapotranspiration on ESVs (e.g., carbon sequestration, oxygen release, and water conservation) in rubber plantations, a multifactorial analytical framework is essential. Our research uses quantitative and qualitative methods, including multivariate statistical methods and mechanistic modeling, to clarify the complex relationships among the influencing factors and differentiate and quantitatively assess the degree to which climatic factors and rubber plantation expansion contribute to the amount of evapotranspiration. The study has three main objectives, namely: (1) reconstructing actual evapotranspiration (ET_{a_reconstructed}) of rubber plantations and calculating the spatial distribution of evapotranspiration in rubber plantations by using Landsat TM/ETM/OLI data to identify rubber plantations and quantifying reference evapotranspiration (ET₀) and actual crop coefficient (K_c,actual) through the GIS platform,; (2) identifying the drivers’ contribution to the spatial heterogeneity of ET_{a_reconstructed} in rubber plantations by using random forest (RF) models of the study area, and (3) determining the optimal interval of ET_{a_reconstructed} through a coupling model of the economic and ecological values of rubber plantations. Sustainable management strategies based on ET_{a_reconstructed} optimization were explored.

2 Materials and methods

2.1 Study area

Xishuangbanna and Pu’er, Yunnan Province, are located in Southwestern China (N 21°10′–24°50′, E 99°09′–102°19′) at an altitude of 726 m (Figure 1). The observation plot of the selected mature rubber plantation was transformed from tropical monsoon forest, and the average age of the rubber trees was 15 years old. The monsoon climate has a distinct rainy season (May–October), when precipitation exceeds 200 mm/month, and a dry season (November–April), with precipitation of less than 50 mm/month. Fog precipitation in the dry season accounted for 5% of the total annual precipitation.

Figure 1

Figure 1. Experimental site of a rubber plantation (indicated by a solid square; 21°34´10″ N, 101°35´24 ″E) in Xishuangbanna, Yunnan Province, Southwest China. (a) Location in Yunnan, China. (b) Observed site of rubber plantation. (c) Automatic meteorological system stations (WS-BR06, Campbell, USA). (d) Drain gauge G3 (METER, USA).

Rubber trees had a plant spacing of 3 m × 6 m to facilitate rubber cutting, and the width of the reclamation area around the mountains was 1.8–2.5 m. Undergrowth vegetation consisted of a small amount of weeds and shrubs. The rubber tapping period was from April to November. In response to cold stress, rubber trees start dormancy in December, and synchronized defoliation occurs in February. Based on a comprehensive analysis of litterfall production, leaf area index dynamics, canopy color, and sap flow, the rubber plantation’s growing season is categorized into four distinct phenological periods, namely: the initial period (March–April, ~61 days), characterized by leaf expansion and flowering; the rapid growth period (May–June, ~61 days), marked by fruiting and peak water use efficiency; the mid-period (July–September, ~92 days), representing the main growth period with a dense canopy and vigorous physiological activity; and the late period (October–November, ~61 days), during which growth gradually declines in preparation for dormancy. This is followed by a dormant period (December–February, ~90 days) of concentrated leaf fall and physiological inactivity (Liu et al., 2014; Ling et al., 2021; Zhou et al., 2021). The rubber plantation evapotranspiration calculation in this study was confined to the growing season, excluding the dormant period.

The canopy tower, standing 20 m tall, was constructed at the study site for continuous micrometeorological and energy flux measurements from 2016. The latent heat flux (LE) and sensible heat flux (H) were measured using a Bowen ratio energy balance (BREB) system. The system was equipped with paired temperature and relative humidity probes (HMP155A, Vaisala, Finland) installed at different heights to determine the vertical gradients. Supporting measurements included net radiation (CNR4, Kipp & Zonen, Netherlands) and soil heat flux (HFP01SC, Hukseflux, Netherlands). Wind speed was measured with a three-dimensional ultrasonic anemometer (WindMaster Pro, Gill Instruments, UK). All signals were sampled at a frequency of 10 Hz and recorded by a dedicated data logger (CR1000, Campbell Scientific, USA). We used ”Bowenby ratio closure” method for closing the canopy surface energy balance for each 30-min interval. The meteorological data and energy flux data were processed by coordinate correction, wild point rejection, data interpolation, and data quality analysis to eliminate abnormal values of rubber plantation due to uncontrollable factors (Falge et al., 2001; Baldocchi et al., 2001).

2.2 Methods and data

2.2.1 Identification and extraction of rubber plantations

A Landsat TM/ETM/OLI time-series NDVI variation spectral library of rubber plantations was constructed using land use/vegetation cover data from Xishuangbanna and Pu’er. Data was extracted from identified rubber plantations by combining the topography and rubber plantation phenology characteristics in Xishuangbanna during the typical time window (February to March) (Sari et al., 2022; Azizan et al., 2021). Remote sensing data for typical days were selected (February 25, 2000; February 22, 2010; and February 27, 2022).

Rubber plantation identification and extraction can be accomplished through an object-oriented classification method that extracts data hierarchically (Lu et al., 2019). Data from rubber plantations in Xishuangbanna and Pu’er in 2000, 2010, and 2022 were extracted at the optimal segmentation scale of 100, shape index of 0.6, and tightness index of 0.9. Data on the spatial distribution of rubber plantations were also obtained.

2.2.2 Estimation of long-term rubber plantation evapotranspiration (1970–2022)

The estimation of rubber plantation evapotranspiration followed the standardized terminology and framework for evapotranspiration assessment as recommended by DeJonge et al. (2025).

1. Calculation of the reference evapotranspiration (ET₀).

The Penman–Monteith method has been recommended as a standard model to calculate ET₀ according to the Food and Agriculture Organization of the United Nations (FAO) (Allen et al., 1998).

$\begin{array}{ll}E{T}_0=\frac{0.408\Delta \sum {\mathrm{(R}}_n-\mathrm{G)}\sum +\gamma \frac{900}{T+273}{u}_2(e_s-e_a)}{\Delta +\gamma (1+0.34u_2)}& (1)\end{array}$

where ET₀ is the reference crop evapotranspiration, mm/day; G is the soil heat flux, MJ/m²·day; R_n is the net canopy surface radiation, MJ/m²·day; U₂ is the mean wind speed at 2 m (which is converted from the wind speed at 10 m), m/s; T is the mean air temperature, °C; e_a is the actual water vapor pressure, kPa; e_s is the saturated water vapor pressure, kPa; γ is the psychrometric constant, kPa/°C; and δ is the saturated water vapor pressure–temperature slope, kPa/°C.

2. Calculation of the actual crop coefficient (K_c,actual).

We employed the single crop coefficient approach to calculate K_c,actual which represents the integrated effect of both plant transpiration and soil evaporation under actual field conditions, inclusive of any water stress.

Daily actual evapotranspiration (ET_a) for the rubber plantation was directly measured using the Bowen ratio energy balance method during the 3-year period from 2016 to 2018. The corresponding daily ET₀ for the same period was obtained from a nearby meteorological station.

The daily K_c,actual was calculated as the ratio of ET_a to ET₀:

$\begin{array}{ll}{K}_{c,actual}=E{T}_a/E{T}_0& (2)\end{array}$

where a multi-year average K_c,actual was derived. A mean seasonal curve was calculated for each day of the year (DOY) by averaging the K_c,actual values from the 3 years (2016, 2017, and 2018) for that specific DOY. This daily climatological curve was then applied as a fixed seasonal pattern to represent the crop coefficient in the long-term reconstruction.

3. Reconstruction of long-term evapotranspiration (ET_a).

To reconstruct a continuous time series of rubber plantation ET_{a_reconstructed} from 1970 to 2022, the following formula was applied using the long-term ET₀ data for this period:

$\begin{array}{ll}E{T}_{a\_reconstructed,i}=K_{c,actual}\times E{T}_{\mathrm{0,}i}& (3)\end{array}$

where ET_{a_reconstructed,i} is the estimated daily actual evapotranspiration for “i” year in the period 1970–2022, mm/day; K_c,actual is the multi-year average daily crop coefficient derived from the 2016–2018 period [as described in Section 2.2.2 (2)]. ET_0,i is the daily reference evapotranspiration for “i” year in the period 1970–2022, mm/day.

To obtain the spatial distribution of ET_{a_reconstructed} across the rubber plantation in the study area, this study employed an upscaling method based on site calibration and spatial interpolation of meteorological data. Daily K_c,actual values were calculated using Equation 2 in Section 2.2.2, based on site-measured ET_a and ET₀ derived from meteorological observations at a flux tower site representing typical rubber stands. These daily K_c,actual values were then averaged according to phenological stages to derive representative K_c,actual values for the standard phenological periods of rubber trees: initial period, rapid growth period, mid-period, and late period. Meteorological data, including daily T_max, T_min, RH, WD, and SSD were collected from weather stations within and around the study area (shown in Table 1). Daily ET₀ for each station was calculated uniformly using Equation 1. Subsequently, the Kriging spatial interpolation method was applied to generate continuous, spatially distributed daily ET₀ raster maps covering the entire study region.

Table 1

Table 1. Information of national meteorological stations in the study area.

By integrating the abovementioned two steps, the reconstructed daily ETa for each grid cell at location (x, y) was computed using Equation 3: ET_{a_reconstructed}(x, y) = K_c,actual × ET₀(x, y), where ET₀(x, y) represents the Kriging-interpolated reference evapotranspiration value for that grid cell on a given day, and K_c,actual is the crop coefficient assigned according to the corresponding phenological stage of that date. The maximum rubber plantation area was determined as the current level year of rubber plantations in the study area based on the field survey and remote sensing image analysis from 2000 to 2022. Then, we discuss the spatial distribution characteristics of rubber plantations ET_{a_reconstructed,i} under the current level year.

To assess the overall water consumption by rubber plantations at the regional scale, this study calculated the total water consumption volume based on the spatially distributed ET_{a_reconstructed}. The total water consumption is calculated as the product of the ET_{a_reconstructed} of each grid cell and its area. This metric represents the total volume of water resources consumed by the entire rubber plantation area over a specific period. For the whole study area, the total water consumption is obtained by aggregating the water consumption of all rubber plantation grid cells. The correlation analysis between rubber plantation area and ET_{a_reconstructed} in this study essentially refers to the analysis between the rubber plantation area and the resulting total regional water consumption.

According to statistics collected from meteorological stations in the study area, the maximum temperature (T_max) was 27.2°C, the minimum temperature (T_min) was 15.8°C, the annual mean relative humidity (RH) was 79.7, the sunshine duration (SSD) was 5.70 h, the wind speed (WD) was 0.8 m/s, and the precipitation (P) was 1,472 mm. The meteorological elements of the study area (1970–2022) are presented in Table 1. P decreased by about 21 mm per decade. The most significant decrease in P was observed in the city of Jinghong. SSD increased by 0.092/(h·d·decade), and annual mean RH varied at a rate of 0.714 per decade. The minimum temperature increased at a rate of 0.219°C per decade. Pu’er was the area with the most significant increase in minimum temperature. The average temperature, SSD, and average WD showed increasing trends, while the average RH and P decreased.

2.2.3 Analysis of rubber plantation evapotranspiration spatial stability

To assess the spatial pattern of inter-annual stability of evapotranspiration in the rubber plantation, the inter-annual coefficient of variation (CV) was calculated for each grid cell (Ayehu et al., 2020).

$\begin{array}{ll}\text{CV }=\text{ ET}\_\text{sd }/\text{ E}{\text{T}}_{\text{a}\_}\text{avg}& (4)\end{array}$

Based on multi-year annual ET_{a_reconstructed} raster data, two key datasets were derived: ET_a_avg, representing the multi-year mean annual ET_{a_reconstructed} for each grid cell, and ET_sd, representing the standard deviation of annual ET_{a_reconstructed} for each grid cell, mm/year; CV values reflect the degree of variability of ET_{a_reconstructed} in rubber plantations, dimensionless. CV values less than 0.1 indicate that plantations are very stable, values between 0.1 and 0.2 are unstable, values between 0.2 and 0.3 are unstable, and values larger than 0.3 are very unstable (Wang and Dickinson, 2012).

2.2.4 Contribution rate

RF models sample objects and variables to construct prediction models. The classification results of each decision tree were summarized based on the variable importance measure (VIM) in the RF algorithm. RF models were designed to assess the contributions of climate and rubber plantation area indicators to ET_{a_reconstructed} performance and rank the importance of those indicators. The simulation results indicated that the method can accurately and quickly select the indicators that have a greater impact on the evaluation results.

2.2.5 Coupled ecosystem services values: rubber plantation valuation models

1. Ecological value assessment of rubber plantations.

The primary ecological value of rubber plantation ecosystems lies in their water regulation and water purification functions. The water regulation value is calculated using the formula below:

$\begin{array}{ll}{U}_{\text{wr}}=10\times {\text{C}}_r\times \text{A}\times \sum \text{(P}-E{T}_{a\_reconstructed}-R)& (5)\end{array}$

U_wr is annual water regulation value, yuan/year; C_r, is reservoir construction cost per unit storage capacity, yuan/m³, with a value of 6.11 yuan/m³ (Lu et al., 2014); A is rubber plantation area, ha; P is annual precipitation, mm/year; ET_{a_reconstructed} is annual evapotranspiration, mm/year; and R is annual surface runoff, mm/year. Data was obtained from research on the Xishuangbanna Tropical Botanical Garden conducted by the Chinese Academy of Sciences (Zhang et al., 1997).

For water purification, the following formula was used:

$\begin{array}{ll}{\text{U}}_{wp}=10\times \text{K}\times \text{A}\times ({\text{P-ET}}_{\text{a\_reconstructed}}\text{-R})& (6)\end{array}$

U_wp is the annual water purification value, yuan/year, and K is the water treatment cost, yuan/ton, in this case 2.09 yuan/ton (Lu et al., 2014). The remaining variables were defined in Equation 5.

2. Carbon sequestration and oxygen release calculation.

The photosynthetic process reflects the coupling relationship between ecosystem carbon and water cycles:

$\begin{array}{ll}GPP=WUE\times Yield-scaledE{T}_{a\_reconstructed}\times C\_c& (7)\end{array}$

GPP is gross primary productivity, g·C/m²; WUE is water use efficiency, g·C/kg, which was based on data from research on the Xishuangbanna Tropical Botanical Garden carried out by the Chinese Academy of Sciences (Lin et al., 2018); and yield-scaled ET_{a_reconstructed} is crop evapotranspiration, kg/m²·mm. C_c is the conversion factor, representing the carbon content in the economic dry yield, with a value of 0.5 g C g^-¹ dry yield. The WUE of rubber trees represents the grams of dry latex biomass produced per kilogram of water consumed, while the yield-scaled ET_{a_reconstructed} indicates the kilograms of latex yield obtained per square meter of land per millimeter of water consumed.

The stoichiometric analysis shows that 1 kg·C = 2.2 kg organic matter, so the photosynthetic production of 1.00 kg organic matter requires 1.63 kg CO₂ fixation (containing 0.27 kg·C) and releases 1.19 kg O₂. Regional carbon/oxygen fluxes in southwestern Yunnan were calculated from vegetation net primary productivity (NPP) (Wang et al., 2019). NPP is derived by subtracting autotrophic respiration (R_a) from GPP which calculated using Equation 7 (Lin et al., 2018):

$\begin{array}{ll}{\text{W}}_{C{O}_2}=NPP\times 1.2\times 1.63& (8)\end{array}$

$\begin{array}{ll}{W}_C=W_{C{O}_2}\times 0.27& (9)\end{array}$

$\begin{array}{ll}{W}_0=NPP\times 2.2\times 1.19& (10)\end{array}$

W_CO2 represents the amount of CO₂ fixed per unit area, g/m²; W_C denotes the corresponding carbon sequestration per unit area in wetlands, g/m²; and NPP indicates the net primary productivity of wetland vegetation per unit area per year.

Primary productivity is calculated by the conversion of atmospheric CO₂ into oxygen through photosynthesis:

$\begin{array}{ll}{\text{U}}_{{\text{O}}_2}=1.19\times {\text{C}}_{O_2}\times \text{A}\times \text{B}& (11)\end{array}$

U_O2 is the annual oxygen release value of rubber plantations, yuan/year; C_O2 is the market price of oxygen, yuan/ton; A is forest area, ha; and B is annual forest net productivity, tons/(ha·year).

3 Results

3.1 Rubber plantation area changes in Xishuangbanna and Pu’er

The rubber plantation area was 16.86 × 10⁴ ha in 2000, and it increased to 30.17 × 10⁴ ha by 2022. The distribution of rubber plantations in Pu’er was less than the distribution in Xishuangbanna, and it was 1.13 × 10⁴, 4.43 × 10⁴, and 12.8 × 10⁴ ha, respectively. The rubber plantation area increased by about 12.49 × 10⁴ ha/decade. Xishuangbanna had an average annual growth rate of 5,789 ha/year, and the average annual growth rate increased from 2000 to 2022. The largest growth rate (10,720 ha/year) was seen from 2000 to 2010. After 2000, the planted areas of rubber plantations in Pu’er expanded rapidly, with an average annual growth rate of 76,108 ha/year. The average annual growth rate in Pu’er also increased between 2000 and 2022 (Figure 2).

Figure 2

Figure 2. Growth trend of rubber plantation area and rubber price in Xishuangbanna and Pu’er (a) 2000; (b) 2010; (c) 2022; (d) price and rubber plantation area trend (data from Yunnan Provincial Bureau of Statistics).

Since 2000, natural rubber prices have exhibited a pattern of an initial increase followed by a decline. It peaked in 2011 with an annual average price of 34,000 yuan/ton. Coinciding with this price maximum, rubber plantation expansion reached its zenith in 2011; the planted area increased by 56.3% during the 2000–2011 period. Subsequently, rubber prices entered a prolonged downward trend and declined to an annual average of 13,500 yuan/ton in 2014, which represented a 60.2% reduction from the peak price. Concurrently, the total rubber plantation area stabilized within the range of 38.67 × 10⁴–43.27 × 10⁴ ha.

3.2 Spatiotemporal variation of rubber plantation evapotranspiration

The K_c,actual of rubber plantations calculated using Equation 2 (Figure 3) for the growing period ranges from 0.72 to 1.49. In the initial period, K_c,actual is 0.89, mid-period is 1.10, and late period is 0.91.

Figure 3

Figure 3. Crop coefficient (K_c,actual) of rubber plantation.

In the last 53 years (1970–2022), the maximum value of rubber plantation ET_{a_reconstructed} was 1,098 mm (in 2020), and the minimum was 871 mm in 1990. The multi-year mean was 952 mm. Based on a linear regression, the mean increased at a rate of 12 mm/decade, but the increasing trend was not significant (Figure 4a). An MK test showed that the ET_{a_reconstructed} from rubber plantations decreased from 1980 to 1990 and increased after 1990. The mutation point around the year of 1995 is shown in Figure 4b.

Figure 4

Figure 4. Interannual variation of ET_{a_reconstructed} and M-K variation trend analysis of rubber plantations in Xishuangbanna. (a) ET_{a_reconstructed} from 1970 to 2022; (b) M-K variation trend analysis from 1970 to 2022.

The spatial distribution of the multi-year average rubber plantation ET_{a_reconstructed} in Xishuangbanna and Pu’er, obtained by using Equation 3 for the layer operation on ArcGIS 10.5, varied significantly from 1970 to 2022 (53 years) (Figure 5a). It was higher in the west and lower in the east: the central and western areas of Menghai county, Lancang county, and Jinghong county were high (annual average ET_{a_reconstructed} >900 mm), while the eastern areas of Mengla county and Jiangcheng county had lower distributions that ranged from 827 to 900 mm.

Figure 5

Figure 5. (a, b) Spatial distribution and stability of annual ET_{a_reconstructed} of rubber plantations.

The CV of rubber plantation ET_{a_reconstructed}, calculated using Equation 4, ranged from 0.04 to 0.11 over the 53 years, which indicates that the degree of variation in rubber plantation ET_{a_reconstructed} was stable (Figure 5b) throughout the study area. The CV values were<0.1 in the rubber plantations in Menghai county, the western areas of Xishuangbanna, Jinggu county, and Mojiang county in the north and east part of Pu’er, which indicates a very low interannual variability of ET_{a_reconstructed} in the rubber plantations in these areas. CV values of >0.1 showed that variation was slightly higher in Lancang county, Jinghong county, Mengla county, and Jiangcheng county. The interannual variation of ET_{a_reconstructed} was relatively unstable, likely due to variability in precipitation and surface moisture conditions.

3.3 Correlation analysis between evapotranspiration and driving factors

Meteorological variables (i.e., average T_min, T_max, P, SSD, WD, RH, and rubber plantation area) were selected for correlation analysis with the annual average ET_{a_reconstructed} in the rubber plantations. Annual ET_{a_reconstructed} in the study area was significantly positively correlated with T_max (correlation coefficients ranged from 0.23 to 0.8) and SSD (0.84–0.93). ET_{a_reconstructed} was significantly negatively correlated with RH (correlation coefficients of -0.36 to -0.75) (Figure 6) and P (-0.03 – -0.49). More precipitation results in a lower ET_{a_reconstructed} because it leads to a higher RH and a lower SSD. Greater WD resulted in an increased ET_{a_reconstructed}. Rubber plantation area showed a positive correlation with the regional total water consumption derived from these plantations (correlation coefficients ranging from 0.1 to 0.75), while a negative correlation was observed with the regionally averaged RH (-0.02 to -0.63). This indicates that the expansion of rubber plantations directly leads to an increase in the total volume of water resources consumed by this ecosystem at the watershed or regional scale, thereby exerting a significant impact on the balance of regional water resources.

Figure 6

Figure 6. Correlation analysis of evapotranspiration driving factors in Xishuangbanna and Pu’er rubber plantations. (a) Mengla, (b) Jinghong, (c) Menghai, (d) Lancang, (e) Jiangcheng, (f) Pu’er, (g) Ning’er, (h) Jinggu, and (i) Mojiang. Area (r) is area of rubber plantations; ET_a is ET_{a_reconstructed}. *represent the P<0.05 as "statistically significant difference", ** represent the P<0.01 as "highly statistically significant difference".

3.4 Effect of meteorological factors and planted area on the spatiotemporal distribution of rubber plantation evapotranspiration

The ET_{a_reconstructed} increased in the central and southeastern regions of our study area (mainly in Jinghong county and Mengla county) and decreased in the west and northeast regions (Mojiang county and Jiangcheng county). Through the RF model, we identified the driving factors of ET_{a_reconstructed} variation in rubber plantations in Xishuangbanna and Pu’er.

Based on the RF, T_min contributed 4.15%–15.33% of the change in ET_{a_reconstructed}. T_min had a higher impact in central Xishuangbanna, Lancang county, and Pu’er county and a lower impact in the southeastern area. T_max contributed 12.33%–19.48% of the change in ET_{a_reconstructed}. The high-value areas were distributed in Mengla county (in southeastern Xishuangbanna) and Pu’er county (central Pu’er). WD contributed 2.78%–16.52% of the change in ET_{a_reconstructed}, and the effect was greater in Xishuangbanna than in Pu’er. RH had a contribution of 11.40%–19.21%, and it was lower in the western part of the study area and higher in Mengla county. P had a contribution of 3.95%–6.76%, which was the lowest factor that affected the change in ET_{a_reconstructed}. SSD and rubber plantation area were the two key drivers of the change in ET_{a_reconstructed}, with contributions of 28.50%–38.42% (SSD) and 11.43%–20.33% (area). SSD had a greater impact on ET_{a_reconstructed} in the southeastern Mengla county and Jiangcheng county than in the other regions. Rubber plantation area had a greater impact in Jinghong county, Menghai county, Jinggu county, and Lancang county (i.e., the western region of the study area) than in the other regions (as shown in Figure 7 (a)-(g)).

Figure 7

Figure 7. Driving factor contribution of evapotranspiration variation of rubber plantations. (a) T_min, (b) T_max, (c) WD, (d) SSD, (e) RH, (f) P, and (g) rubber plantation area.

3.5 Influence of rubber plantation evapotranspiration on ecological and economic factors

Rubber plantation ET_{a_reconstructed} directly reflects the WUE, which is closely associated with ecological service functions, such as carbon sequestration, oxygen release, and water conservation. During the period from 2020 to 2022, rubber plantation ecosystem WUE was lower in the rainy season (2.60 ± 1.22 g·kg^-¹) than in the dry season (1.52 ± 0.70 g·kg^-¹).

We employed multiple valuation approaches, including the shadow project method, market value method, and cost analysis method, to quantify the comprehensive ESVs of rubber plantations. The valuation results are presented in Table 2. (calculated by Equation 5–11)

Table 2

Table 2. Ecosystem service values of rubber plantations in different ET_{a_reconstructed}.

The rubber plantation's economic value of the conservation value of water resources and carbon sequestration and oxygen release (as shown in Figure 8). Figure 9 elucidates the trade-offs and synergies between the hydrological cycle and the eco-economic service values in rubber plantation ecosystems. Figure 9a reveals a critical trade-off between the carbon sequestration and oxygen release value and the water conservation value in relation to ET_{a_reconstructed}. A non-linear, positive correlation was observed between ET_{a_reconstructed} and carbon sequestration value, with the highest eco-economic efficiency occurring within an optimal ET_{a_reconstructed} range of 2 to 3 mm mm/day. Specifically, for every 10% increase in ET_{a_reconstructed} within this range, the economic value of carbon sequestration and oxygen release increased by approximately 6 × 10⁵ yuan. However, this economic gain comes at a cost to water resources. The water conservation value exhibited a significant decline with rising ET_{a_reconstructed}. This negative impact was 1.4 times more pronounced during the dry season, a period when ET_{a_reconstructed} frequently exceeds precipitation, thereby exacerbating water scarcity and posing a threat to regional water security.

Figure 8

Figure 8. Economic value of the conservation value of water resources and carbon sequestration and oxygen release.

Figure 9

Figure 9. Ecological and economic effects of rubber plantations. Evapotranspiration and the economic value of carbon sequestration and oxygen release as well as the conservation value of water resources (a) SSD and carbon sequestration and oxygen release value (b) ET_a is ET_{a_reconstructed}.

Figure 9b illustrates the synergistic mechanism driven by SSD. As the dominant meteorological factor controlling ET_{a_reconstructed}, SSD regulates photosynthetically active radiation, thereby directly enhancing GPP and, consequently, the carbon sequestration value. The analysis identifies a multi-year average SSD of 4.17 h as an optimal threshold that corresponds to the peak carbon sequestration and oxygen release value in rubber plantations. This figure demonstrates that while climatic drivers like SSD can synergistically promote both transpiration and carbon uptake, the resulting high water consumption creates a direct conflict between maximizing carbon-related economic benefits and ensuring sustainable water resource conservation.

4 Discussion

4.1 Expanded planted areas on rubber plantations

Our results demonstrated the substantial expansion of rubber plantations in Xishuangbanna and Pu’er, which increased by approximately 12.49 × 10⁴ ha per decade from 2000 to 2022. This result represents a profound land use transformation primarily driven by market forces: the expansion trajectory closely tracked price fluctuations in the natural rubber market. The most rapid phase of expansion occurred during 2000–2011 and peaked in 2011 when rubber prices reached their maximum (34,000 yuan/ton), which corresponded to a 56.3% increase in planted area. This synchronous pattern strongly indicates that favorable market conditions served as the primary catalyst for initial land conversion. The subsequent price decline to 13,500 yuan/ton by 2014, which represented a 60.2% reduction from the peak, coincided with a notable stabilization of plantation area and suggested a strategic shift from extensification to intensification management as economic incentives diminished.

This market-driven expansion had profound ecohydrological implications. The replacement of native ecosystems with rubber monocultures, which are known for their higher evapotranspiration, substantially altered regional water cycles (Lin et al., 2022; Gnanamoorthy et al., 2022). From local to regional scales, the impact of rubber plantation expansion on the hydrological cycle has increased in the humid tropics—for example, a 50% increase in rubber plantation area increased evapotranspiration by 3.3% (Celine and James, 2015). Fog precipitation, a significant water resource, decreased because of LUCC in Xishuangbanna (Peng et al., 2020). The climate is gradually changing from humid to dry, which is accelerating the formation of climate extremes (Chen et al., 2024) and causing negative ecohydrological effects in the form of regional water resource shortages (Ziegler et al., 2009; Guardiola-Claramonte et al., 2010).

4.2 Spatiotemporal distribution of evapotranspiration

The multi-year rubber plantation ET_{a_reconstructed} in our study was 952 mm. We cross-validated the model outputs against field measurements obtained using the Bowen ratio–energy balance method in the study area (Ling et al., 2022b). The results demonstrate a strong agreement in seasonal variation trends during the growing season, with an annual relative error of 6%. Carr (2012) found an evapotranspiration value of 1,050 mm in rubber plantations in Thailand. Tan et al. (2011) suggested that evapotranspiration on rubber plantations was 1,137 mm based on catchment water balances and 1,125 mm based on eddy covariance in Xishuangbanna. The slight underestimation in our results may be attributed to local climatic conditions, such as higher relative humidity, reduced sunshine hours, and the age structure of the rubber trees. This observed discrepancy is considered reasonable, and it is consistent with regional characteristics (Qiao et al., 2025).

Rubber plantations exhibit higher evapotranspiration and greater water consumption than secondary forests (Igarashi et al., 2015). In Thailand, the actual water demand of newly established rubber plantations has been reported to exceed twice the water demand of conventional crops (Mangmeechai, 2020). In this study, the spatial interpolation results showed a deviation of less than 16% from the field measurements obtained at the Xishuangbanna Rubber Meteorological Center, which indicates an acceptable accuracy of the spatially distributed ET_{a_reconstructed} estimates. The uneven distribution of precipitation and radiant energy in the region likely led to spatial differences in ET_{a_reconstructed}.

Rubber plantations exhibit higher water consumption than the prevailing crops in Southeast Asia (Chiarelli et al., 2018; Pradeep et al., 2022). Our findings from southwestern China are consistent with this pattern. They reveal significant, increasing trends in evapotranspiration from rubber plantations over the past 53 years, with a notable abrupt change that occurred around the year 1995. The rapid expansion of rubber plantations exhibited a temporal trend consistent with the region’s aridification. Large-scale rubber cultivation has contributed significantly to local temperature increases and relative humidity reduction. Concurrently, southwestern China experienced an abrupt shift in potential evapotranspiration around 1995, accompanied by mutated extreme precipitation patterns and increased drought frequency (Liu and Yang, 2021; Lang et al., 2017).

4.3 Factors that influence rubber evapotranspiration

The driving factors for ET_{a_reconstructed} were selected based on their fundamental roles in the surface energy balance and water vapor transfer processes that govern evapotranspiration. T_min and T_max are proxies for the thermal energy available for evaporation, representing the water supply. SSD is a direct indicator of solar radiation, the dominant energy source for evapotranspiration. WD influences turbulent transport and affects the removal of water vapor from the surface. RH determines the vapor pressure deficit. Finally, rubber plantation area was included as a crucial land use variable to quantify anthropogenic alterations to the land surface.

The expansion of rubber plantations directly increases regional ET_{a_reconstructed} by modifying the key biophysical properties of the land surface. Rubber trees allow a greater interception of solar radiation and access deeper soil moisture, which sustain elevated transpiration rates (Ling et al., 2022a). This intensified water use establishes a local climate feedback mechanism whereby increased atmospheric water vapor from transpiration lowers the near-surface RH. This finding was supported by the significant negative correlation observed between rubber plantation area and RH. The resulting drier air amplifies atmospheric evaporative demand, and the regional hydrological balance shifts as a larger proportion of precipitation is allocated to ET_{a_reconstructed}. Reduced streamflow and groundwater recharge intensify water scarcity during dry periods (Perron et al., 2024; Gnanamoorthy et al., 2022). This outcome is consistent with previous studies reporting substantial runoff reductions in watersheds following rubber plantation establishment.

4.4 Strategies to enhance ecosystem services values through rubber plantation evapotranspiration management

The rapid expansion of rubber plantations in Southeast Asia has increased water consumption—for example, green water consumption increased by 38 km³/year (Chen et al., 2024). ET_{a_reconstructed} is a critical influence on the ESVs of forest–grassland ecosystems. In semi-arid mountainous regions, potential evapotranspiration serves as a key factor that determines carbon sequestration capacity (Wang, 2024). Based on the soil–plant–atmosphere continuum theory, the photosynthetic–evapotranspiration coupling model in winter wheat validated the diurnal relationship between evapotranspiration and CO₂ flux (Wang et al., 2004). Recent research has analyzed the decoupling phenomenon between evapotranspiration and carbon sequestration under drought stress using the Shuttleworth–Wallace dual-source model and the EALCO model. The results highlighted the impact of seasonal water deficits on WUE (Bi, 2023).

The ecological functionality of rubber plantations is only 21.5%–63.6% of that of primary forests (Singh et al., 2021), which suggests that optimized evapotranspiration management can enhance their ESVs. In high-evapotranspiration zones, such as the northwestern regions of Xishuangbanna and Pu’er, carbon sequestration and oxygen release values are elevated. Excessive WUE may induce water stress, impair physiological activity, and reduce water conservation capacity, which results in seasonal water scarcity. Conversely, low-evapotranspiration zones, such as the southeastern regions, exhibit diminished WUE, which limits photosynthesis and weakens carbon sequestration (Lin et al., 2022). Evapotranspiration critically influences rubber yield and economic returns. Excessive evapotranspiration under prolonged drought conditions reduces bark turgor pressure, which hinders latex flow and decreases tapping productivity. Insufficient evapotranspiration during the waterlogged rainy seasons may cause root hypoxia, which also suppresses latex production. SSD is positively correlated with evapotranspiration. High SSD enhances evapotranspiration but accelerates soil moisture depletion, which triggers stomatal closure and WUE reduction (Zeng et al., 2022). Thus, agroforestry intercropping with moderate shading can mitigate water stress. Rubber–Flemingia macrophylla intercropping systems in high-evapotranspiration areas can reduce the evapotranspiration caused by a monoculture while improving overall WUE and balancing water consumption and economic gains (Liu et al., 2019).

4.5 Limitations and future perspectives

While this study elucidates the spatiotemporal patterns of ET_{a_reconstructed} in rubber plantations and its impact on ESVs, its focus on southwestern China (Xishuangbanna and Pu’er) may constrain its generalizability. This limitation arises from the significant heterogeneity in climate, topography, and management practices among Southeast Asia’s principal rubber-growing regions. Extending the proposed methodology to broader geographical contexts would help verify its applicability and reveal regional differentiation patterns.

Furthermore, this study established a baseline average for K_c,actual of rubber plantations based on field measurements from 2016 to 2018. While using a 3-year average partially mitigates interannual variability, applying this value as a constant for reconstructing a 53-year historical series risks overlooking substantial influences from climatic decadal variability and vegetation life cycle succession. Therefore, we emphasize the representativeness limitations of this averaged value and recommend that future studies utilize such multi-year observational datasets as key benchmarks to calibrate eco-hydrological models capable of dynamically simulating crop coefficients and soil water stress coefficient, thereby enabling more reliable historical reconstruction and future prediction.

5 Conclusion

Our research investigated the impacts of climate change and rubber plantation expansion on ET_{a_reconstructed} and its main implications for ESVs in Southwestern China. The spatial distribution of ET_{a_reconstructed} was characterized by higher values in the western regions of the study area (968 mm) and lower values in the eastern regions (828 mm). The developed ET_{a_reconstructed} model successfully facilitated the scaling of estimations from site-level observations to a regional scale. The expansion of rubber plantations was identified as a significant driver of ET_{a_reconstructed} that contributed to 11.4%–20.3% increase in regional ET_{a_reconstructed}. This increase was primarily driven by climatic factors, of which increasing SSD and RH were the most influential. An optimal ET_{a_reconstructed} range of 2 to 3 mm/day was established for rubber plantations, which achieves a balance among water consumption, carbon sequestration, and latex yield. ET_{a_reconstructed} rates outside this range result in either water stress or reduced photosynthetic productivity. The optimal rubber plantation ESVs were achieved when the multi-year average SSD was 4.17 h.

In conclusion, this research provides a critical scientific basis for sustainable water resource management and climate-resilient planning in rubber-producing regions. It highlights the necessity of regulating plantation density and water use to optimize eco-economic benefits.

Data availability statement

Publicly available datasets were analyzed in this study. All spatial dataset files are available from Resource and Environment Science and Data Center (URL: https://www.resdc.cn/DOI/DOI.aspx?DOIID=120).

Author contributions

ZL: Conceptualization, Data curation, Investigation, Methodology, Writing – original draft. TX: Funding acquisition, Writing – review & editing. YS: Resources, Writing – review & editing. GH: Data curation, Methodology, Writing – review & editing. SG: Supervision, Writing – review & editing. ZZ: Investigation, Resources, Writing – review & editing. BC: Investigation, Resources, Writing – review & editing.

Funding

The author(s) declare that financial support was received for the research and/or publication of this article. National Key R&D Program for the 14th Five-Year Plan (no. 2021YFC3000205-06), demonstration project of comprehensive government management and large-scale industrial application of the major special project of CHEOS (no. 89-Y50G31-9001-22/23-05), Reserve Candidates for Young and Middle-aged Academic and Technical Leaders in Kunming (202405C040060), Frontier Research Team of Kunming University 2023, Special Basic Cooperative Research Programs of Yunnan Provincial Undergraduate Universities Association (202301AO070165) (202401BA070001-099), Yunnan Province Young and middle-aged Academic and Technical Leaders Reserve Talent Program (202405AC350040), Yunnan Provincial Natural Ecological Monitoring Network Monitoring Project (2024-YN-18), and the Scientific Research and Technical Innovation Team Construction of Yunnan Province (no. 2018HC024), China.

Conflict of interest

Authors ZZ and BC were employed by the company Yunnan Tobacco Company.

The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Generative AI statement

The author(s) declare that no Generative AI was used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

References

Allen R. G., Pereira L. S., Raes D., and Smith M. (1998). “Crop evapotranspiration: Guidelines for computing crop water requirements,” in FAO irrigation and drainage paper no. 56 (Food and Agriculture Organization of the United Nations, Rome).

Google Scholar

Ayehu G., Tadesse T., and Gessesse B. (2020). Monitoring residual soil moisture and its association to the long-term variability of rainfall over the upper Blue Nile Basin in Ethiopia. Remote Sens. 12, 2138. doi: 10.3390/rs12132138

Crossref Full Text | Google Scholar

Azizan F. A., Astuti I. S., Aditya M. I., Febbiyanti T. R., Williams A., Young A., et al. (2021). Using multi-temporal satellite data to analyse phenological responses of rubber (Hevea brasiliensis) to climatic variations in South Sumatra, Indonesia. Remote Sens. 13, 2932. doi: 10.3390/rs13152932

Crossref Full Text | Google Scholar

Baldocchi D., Falge E., Gu L., Olson R., Hollinger D., Running S., et al. (2001). FLUXNET: A new tool to study the temporal and spatial variability of ecosystem-scale carbon dioxide, water vapor, and energy flux densities. Bull. Am. Meteorological Soc. 82, 2415–2434. doi: 10.1175/1520-0477(2001)082<2415:FANTTS>2.3.CO;2

Crossref Full Text | Google Scholar

Bi Z. Y. (2023). Estimation of global potential evapotranspiration and its variation attribution analysis based on Shuttleworth-Wallace’s ‘Dual Source’ model (Master’s thesis) Nanjing. (Nanjing University of Information Science & Technology).

Google Scholar

Carr M. K. (2012). The water relations of rubber (Hevea brasiliensis): A review. Exp. Agric. 48, 176–193. doi: 10.1017/S0014479711000901

Crossref Full Text | Google Scholar

Celine G. and James E. J. (2015). Assessing the implications of extension of rubber plantation on the hydrology of humid tropical river basin. Int. J. Environ. Res. 9, 841–852. doi: 10.22059/IJER.2015.967

Crossref Full Text | Google Scholar

Chen Q. Y., Fu R. Y., Cheng S. Y., Qiao D., Hu Z. M., Zhang Z. J., et al. (2024). Effects of the conversion of natural tropical rainforest to monoculture rubber plantations on soil hydrological processes. J. Plant Ecol. 17, rtae021. doi: 10.1093/jpe/rtae021

Crossref Full Text | Google Scholar

Chiarelli D. D., Rosa L., Rulli M. C., and D’Odorico P. (2018). The water-land-food nexus of natural rubber production. J. Cleaner Production 172, 1739–1747. doi: 10.1016/j.jclepro.2017.12.021

Crossref Full Text | Google Scholar

Cutting N. G., Kaur S., Singh M. C., Biswas S., and Mall R. K. (2024). Estimating crop evapotranspiration in data-scarce regions: A comparative analysis of eddy covariance, empirical and remote-sensing approaches. Water Conserv. Sci. Eng. 9, 65. doi: 10.1007/s41101-024-00270-y

Crossref Full Text | Google Scholar

DeJonge K. C., Allen R. G., Kilic A., Thorp K. R., Kukal M. S., Marek G., et al. (2025). Evapotranspiration terminology and definitions. J. Irrigation Drainage Eng. 151, 06025003. doi: 10.1061/JIDEDH.IRENG-10491

Crossref Full Text | Google Scholar

Falge E., Baldocchi D., Olson R., Anthoni P., Aubinet M., Bernhofer C., et al. (2001). Gap filling strategies for defensible annual sums of net ecosystem exchange. Agric. For. Meteorology 107, 43–69. doi: 10.1016/S0168-1923(00)00225-2

Crossref Full Text | Google Scholar

Giambelluca T. W., Mudd R. G., Liu W., Ziegler A. D., Kobayashi N., Kumagai T., et al. (2016). Evapotranspiration of rubber (Hevea brasiliensis) cultivated at two plantation sites in Southeast Asia. Water Resour. Res. 52, 660–679. doi: 10.1002/2015WR017964

Crossref Full Text | Google Scholar

Gnanamoorthy P., Song Q., Zhao J., Zhang Y., Zhang J., Lin Y., et al. (2022). Seasonal fog enhances crop water productivity in a tropical rubber plantation. J. Hydrology 611, 128016. doi: 10.1016/j.jhydrol.2022.128016

Crossref Full Text | Google Scholar

Guardiola-Claramonte M., Troch P. A., Ziegler A. D., Giambelluca T. W., Durcik M., Vogler J. B., et al. (2010). Hydrologic effects of the expansion of rubber (Hevea brasiliensis) in a tropical catchment. Ecohydrology 3, 306–314. doi: 10.1002/eco.110

Crossref Full Text | Google Scholar

Igarashi Y., Katul G. G., Kumagai T., Yoshifuji N., Sato T., Tanaka N., et al. (2015). Separating physical and biological controls on long-term evapotranspiration fluctuations in a tropical deciduous forest subjected to monsoonal rainfall. J. Geophysical Research: Biogeosciences 120, 1262–1278. doi: 10.1002/2015JG002935

Crossref Full Text | Google Scholar

Kumagai T., Mudd R. G., Giambelluca T. W., Kobayashi N., Miyazawa Y., Lim T. K., et al. (2015). How do rubber (Hevea brasiliensis) plantations behave under seasonal water stress in northeastern Thailand and central Cambodia? Agric. For. Meteorology 213, 10–22. doi: 10.1016/j.agrformet.2015.06.011

Crossref Full Text | Google Scholar

Lam S. K., Suter H., Mosier A. R., and Chen D. (2022). Mitigating soil greenhouse gas emissions from land-use change in tropical peatlands. Front. Ecol. Environ. 20, 352–360. doi: 10.1002/fee.2500

Crossref Full Text | Google Scholar

Lang D. X., Shi J. Q., Zheng J. K., Li Y., Yang Y., Wang Y. L., et al. (2017). Temporal variation characteristics of potential evaporation in Southwest China in recent 52 years. Resour. Environ. Yangtze Basin 26, 945–954. doi: 10.11870/cjlyzyyhj201706017

Crossref Full Text | Google Scholar

Li W., Liang Y., Liu L., He Q., Huang J., and Yin Z. (2024). Spatio-temporal impacts of land use change on water-energy-food nexus carbon emissions in China 2011–2020.. Environ. Impact Assess. Rev. 105, 107436. doi: 10.1016/j.eiar.2024.107436

Crossref Full Text | Google Scholar

Lin Y., Zhang Y., Zhao W., Zhou L., Li J., Fei X.-H., et al. (2022). Hail event rather than cold stress, has profound effects on the carbon-water coupling process of Hevea brasiliensis in Xishuangbanna. Ind. Crops Products. 175, 114237. doi: 10.1016/j.indcrop.2021.114237

Crossref Full Text | Google Scholar

Lin Y. X., Grace J., Zhao W., Dong Y. X., Zhang X., Zhou L. G., et al. (2018). Water-use efficiency and its relationship with environmental and biological factors in a rubber plantation. J. Hydrology 563, 273–282. doi: 10.1016/j.jhydrol.2018.05.068

Crossref Full Text | Google Scholar

Ling Z., Shi Z. T., Gu S., He G., Liu X., Wang T., et al. (2022a). Estimation of applicability of soil model for rubber (Hevea brasiliensis) plantations in Xishuangbanna, Southwest China. Water 14, 295. doi: 10.3390/w14030295

Crossref Full Text | Google Scholar

Ling Z., Shi Z. T., Gu S. X., Peng H. Y., Feng G. J., and Huo H. (2021). Energy balance and evapotranspiration characteristics of rubber (Hevea brasiliensis) plantations in Xishuangbanna, Southwest of China. Appl. Ecol. Environ. Res. 20, 103–117. doi: 10.15666/aeer/2001_103117

Crossref Full Text | Google Scholar

Ling Z., Shi Z. T., Gu S. X., Wang T., Zhu W. W., and Feng G. J. (2022b). Impact of climate change and rubber (Hevea brasiliensis) plantation expansion on reference evapotranspiration in Xishuangbanna, Southwest China. Front. Plant Sci. 13. doi: 10.3389/fpls.2022.830519

PubMed Abstract | Crossref Full Text | Google Scholar

Liu W. J., Li J., Liu H., Yang G. R., He H., Huang J. X., et al. (2014). Vertical patterns of soil water acquisition by non-native rubber trees (Hevea brasiliensis) in Xishuangbanna, Southwest China. Ecohydrology 7, 1234–1244. doi: 10.1002/eco.1450

Crossref Full Text | Google Scholar

Liu W. J., Luo Q. P., Li J. T., Wang P. Y., Lu H. J., Liu W. Y., et al. (2015). The effects of conversion of tropical rainforest to rubber plantation on splash erosion in Xishuangbanna, SW China. Hydrology Res. 46, 168–174. doi: 10.2166/nh.2013.109

Crossref Full Text | Google Scholar

Liu C.-A., Nie Y., Rao X., Tang J.-W., and Siddique K. H. M. (2019). The effects of introducing Flemingia macrophylla to rubber plantations on soil water content and exchangeable cations. Catena 172, 480–487. doi: 10.1016/j.catena.2018.08.038

Crossref Full Text | Google Scholar

Liu W. and Yang Y. D. (2021). Reconstruction and analysis of extreme drought and flood events in Southwest China in the past 600 years. Quaternary Sci. 41, 368–378. doi: 10.11928/j.issn.1001-7410.2021.02.06

Crossref Full Text | Google Scholar

Lu H. L., Niu A. Y., Zhao J. C., Li X. W., and Liu X. J. (2014). Evaluation of ecosystem service value in Xishuangbanna area based on land use change. Jiangsu Agric. Sci. 42, 278–281. doi: 10.3969/j.issn.1002-1302.2014.05.102

Crossref Full Text | Google Scholar

Lu M., Zhang L., and Wu S. (2019). Interannual variation trend and cause analysis of reference evapotranspiration in Jialing River Basin. Res. Soil Water Conserv. 26, 77–85. doi: 10.13869/j.cnki.rswc.2019.03.011

Crossref Full Text | Google Scholar

Mangmeechai A. (2020). Effects of rubber plantation policy on water resources and land use change in the northeastern region of Thailand. Geography Environment Sustainability 13, 73–83. doi: 10.24057/2071-9388-2019-118

Crossref Full Text | Google Scholar

Peng H. Y., Shi Z. T., and Tong S. Y. (2020). Climatological characteristics and influencing factors of fog in Xishuangbanna area. Geographical Res. 39, 1907–1919. doi: 10.11821/dlyj020190615

Crossref Full Text | Google Scholar

Perron T., Legrand M., Janeau J. L., Manizan A., Vierling C., Kouakou A., et al. (2024). Runoff and soil loss are drastically decreased in a rubber plantation combining the spreading of logging residues with a legume cover. Sci. Total Environ. 913, 169335. doi: 10.1016/j.scitotenv.2023.169335

PubMed Abstract | Crossref Full Text | Google Scholar

Pradeep B., Sylas V. P., and Jessy M. D. (2022). A framework for assessing the vulnerability of rubber plantations to the impacts of climate change with special reference to Kerala, India. J. Rubber Res. 25, 387–399. doi: 10.1007/s42464-022-00183-4

Crossref Full Text | Google Scholar

Qi D., Yang C., Yun T., and Wu Z. (2023). The main service functions and driving forces of rubber (Hevea brasiliensis) plantation ecosystem in China. J. Rubber Res. 26, 155–164. doi: 10.1007/s42464-023-00202-w

Crossref Full Text | Google Scholar

Qiao S., Liu J., Xu T., and Zhang D. (2025). Livelihood capital, speculative preferences, and natural rubber farmers’ participation in cooperatives. Agriculture 15, 562. doi: 10.3390/agriculture15040562

Crossref Full Text | Google Scholar

Sarathchandra C., Alemu A. Y., Worthy F. R., Lakmali W. I., Ma H. X., Bi Y. F., et al. (2021). Impact of land use and land cover changes on carbon storage in rubber dominated tropical Xishuangbanna, South West China. Ecosystem Health Sustainability 7, 1915183. doi: 10.1080/20964129.2021.1915183

Crossref Full Text | Google Scholar

Sari I. L., Weston C. J., Newnham G. J., and Volkova L. (2022). Developing multi-source indices to discriminate between native tropical forests, oil palm and rubber plantations in Indonesia. Remote Sens. 14, 3. doi: 10.3390/rs14010003

Crossref Full Text | Google Scholar

Singh A. K., Liu W., Zakari S., Wu J., Yang B., Jiang X. J., et al. (2021). A global review of rubber plantations: Impacts on ecosystem functions, mitigations, future directions, and policies for sustainable cultivation. Sci. Total Environ. 796, 148948. doi: 10.1016/j.scitotenv.2021.148948

PubMed Abstract | Crossref Full Text | Google Scholar

Tan Z., Zhang Y., Song Q., Liu W., Deng X., Tang J., et al. (2011). Rubber plantations act as water pumps in tropical China. Geophysical Res. Lett. 38, L24406. doi: 10.1029/2011GL050006

Crossref Full Text | Google Scholar

Thellmann K., Golbon R., Cotter M., Cadisch G., and Asch F. (2019). Assessing hydrological ecosystem services in a rubber-dominated watershed under scenarios of land use and climate change. Forests 10, 176. doi: 10.3390/f10020176

Crossref Full Text | Google Scholar

Wang Y. (2024). Research on carbon flux characteristics and influencing factors of typical forest and grassland ecosystems in mountainous areas of semi-arid regions Lanzhou (Lanzhou University).

Google Scholar

Wang K. and Dickinson R. E. (2012). A review of global terrestrial evapotranspiration: Observation, modeling, climatology, and climatic variability. Rev. Geophysics 50, RG2005. doi: 10.1029/2011RG000373

Crossref Full Text | Google Scholar

Wang J. M., Wu S. Q., Cui P., and Lei J. C. (2019). Assessment on forest ecosystem service value in Wuma River Basin. J. Guizhou Normal University Natural Sci. 37, 49–55. doi: 10.16614/j.gznuj.zrb.2019.04.007

Crossref Full Text | Google Scholar

Wang J., Yu Q., Li X. G., Sun X. M., and Zhu Z. L. (2004). Simulation of diurnal variation of CO2 flux in winter wheat using photosynthesis-evapotranspiration coupling model. Acta Ecologica Sin. 24, 2782–2788.

Google Scholar

Wang Y. and Zhang L. (2025). From primary forests to rubber plantations: A huge ecological loss. Innovation 6, 100836. doi: 10.1016/j.xinn.2025.100836

PubMed Abstract | Crossref Full Text | Google Scholar

Xue S., Zhu L., He Y., Li D., and Ye C. (2024). Runoff change characteristics and response to climate variability and human activities under a typical basin of natural tropical rainforest converted to monoculture rubber plantations. Forests 15, 1918. doi: 10.3390/f15111918

Crossref Full Text | Google Scholar

Zeng H. H., Wu J. E., Singh A. K., Zhu X. A., Zhang W. J., Hölscher D., et al. (2022). Effect of intercrops complexity on water uptake patterns in rubber plantations: Evidence from sta ble isotopes (C-H-O) analysis. Agriculture Ecosyst. Environ. 338, 108086. doi: 10.1016/j.agee.2022.108086

Crossref Full Text | Google Scholar

Zhang Y. P., Zhang K. Y., Ma Y. X., Liu Y. H., and Liu W. J. (1997). Runoff characteristics in different vegetation coverage areas of Xishuangbanna Tropical Region. J. Soil Erosion Soil Water Conserv. 3, 26–31.

Google Scholar

Zhou Y., Liu S. L., Xie M. M., Sun Y. X., and An Y. (2021). Dynamics of regional vegetation changes under the disturbance of human activities: A case study of Xishuangbanna. Acta Ecologica Sin. 41, 565–574. doi: 10.5846/stxb201912192733

Crossref Full Text | Google Scholar

Ziegler A. D., Fox J. M., and Xu J. (2009). The rubber juggernaut. Science 324, 1024–1025. doi: 10.1126/science.1173833

PubMed Abstract | Crossref Full Text | Google Scholar