Can Forest Management Improve Water Retention Conservation Under Climate Change? A Case Study of the Republic of Korea

Abstract

This study aimed to analyze changes in water retention conservation in response to climate change and forest management strategies and to propose methods for securing sustainable water resources. The KO-G-Dynamic model, a Korean forest growth model, was utilized alongside aboveground and belowground water resource prediction models to evaluate changes in water retention conservation under various climate change scenarios and forest management strategies. The analysis revealed that under climate change and current forest management levels, water retention conservation was projected to reach 37.553 billion tons per year in the 2030s, 38.274 billion tons per year in the 2050s, and 40.306 billion tons per year in the 2080s. Under optimal forest management policies, the water yield and storage were expected to increase to 37.863 billion tons per year in the 2030s, 38.877 billion tons per year in the 2050s, and 41.495 billion tons per year in the 2080s. Notably, watershed-based forest management offers a more practical management unit than conventional legal boundaries, as it reflects hydrological flow and the ecological characteristics of forest environments. Furthermore, the watershed-based forest management scenario demonstrated greater feasibility in securing water resources. This study provides foundational data for climate change adaptation and sustainable forest management and may aid national and local forest planning. The findings underscore the critical role of forest management in mitigating climate change impacts and ensuring long-term water sustainability.

1. Introduction

Global surface temperatures increased by 1.1 °C between 2011 and 2020 compared to the period of 1850–1900, accelerating climate change [1]. This change affects various sectors, making the prediction of future conditions and the identification of adaptation pathways essential [2]. To mitigate climate change, countries have submitted Nationally Determined Contributions (NDCs) with sector-specific reduction targets, emphasizing the role of forests as carbon sinks. However, forests are increasingly impacted by climate change, including droughts and extreme precipitation, which alter their physical, chemical, and biological processes [3]. Consequently, maximizing forest functions—such as carbon sequestration, water resource enhancement, and disaster prevention—through adaptation strategies that reflect their environmental and ecological characteristics is crucial [4].

In the Republic of Korea, the proportion of forest area has gradually declined from 63.9% in 2014 to 63.1% in 2023, and forest distribution is shifting northward due to climate change [5]. The National Institute of Forest Science (NIFoS) projects that under the SSP5-8.5 scenario, total discharge and base flow will decrease by 6.47 billion m³ and 3.36 billion m³, respectively, by 2040, which is expected to intensify water shortages and drought conditions. Additionally, erosion control and greening projects implemented between 1973 and 1987 prioritized fast-growing species, leading to an increase in climate-vulnerable species and higher mortality rates [6,7]. In particular, data from the 5th and 6th National Forest Inventory (NFI) revealed that coniferous species exhibited a mortality rate that was 96% higher than that of other tree species [8,9]. Coniferous forests, which account for 38.8% of the country’s forests, exhibit higher evapotranspiration rates than broadleaf forests, reducing their capacity for water retention [10]. As stand density increases, canopy interception loss, where water evaporates before reaching the ground, also rises, underscoring the importance of forest management [11]. Since belowground storage holds more water than aboveground components, density management can help reduce evapotranspiration losses [12]. The Korea Forest Service recommends maintaining a canopy closure of approximately 70% under the Green Dam project and conducting light thinning every 5–10 years to manage herbaceous plant growth. Additionally, pruning below the tallest dominant branches is advised to improve forest conditions [13].

However, previous studies on forest water resource estimation have used statistical-based predictive models and stand-level analysis, which allow for the evaluation of overall forest water resource trends [11,14,15]. While this approach has the advantage of assessing general water resource tendencies, it has limitations in evaluating resource quantities that reflect forest hydrological characteristics at the basic spatial unit level and in predicting water resources based on climate change and management strategies [16]. Moreover, forest water retention conservation must account for the varying storage capacities of aboveground and belowground components, necessitating separate assessments for each. A watershed-based standard forest unit water retention conservation model is essential to improve predictions of future water resources and support policy decision-making.

Therefore, this study aimed to evaluate changes in water retention conservation under accelerating climate change and assess the impacts of both current forest management practices and policy-based management strategies, with the goal of exploring pathways for securing sustainable water resources [17,18,19,20]. Additionally, it sought to establish standardized forest management units based on watershed-level hydrological characteristics that reflect forest physiognomy. Using these units, this study estimated the water yield under climate and forest management scenarios, contributing to national and regional forest management planning.

2. Materials and Methods

2.1. Study Area

The study area is situated within the forested regions of the Republic of Korea, covering longitudes from 124°54′ to 131°6′ and latitudes from 33°9′ to 38°45′. The watersheds are categorized into major watersheds (21), medium watersheds (117), standard watersheds (850), and catchment areas (35,951). The effective soil depth varies between 0 and 150 cm, depending on the forest type and environmental conditions (Figure 1) [21,22].

Currently, forests in the Republic of Korea cover approximately 62.59% (6,287,325 ha) of the national land, consisting of 38.8% coniferous forests, 33.4% broadleaf forests, and 27.8% mixed forests [23]. These forests suffered severe degradation due to the war from 1950 to 1953. However, a government-led erosion control and greening project from 1973 to 1987 successfully restored them [24]. As a result, the initial stand volume of 11.31 m³/ha increased to 176 m³/ha by 2023. In particular, Gangwon Province shows a notably high stand volume of 196 m³/ha, followed by Gyeongsangnam-do with 190 m³/ha and Gyeongsangbuk-do with 184 m³/ha, all exceeding the national average [23]. Nevertheless, forest aging due to an imbalance in age-class distribution, along with an increase in climate-vulnerable species, is leading to mortality and a decline in water resource functions [25].

This region was selected as the study area to explore strategic approaches for securing sustainable water retention conservation by integrating climate adaptation and forest management, while reflecting the hydrological and ecological characteristics of watersheds in the Republic of Korea.

2.2. Forest Dynamic Stand Growth Model

To simulate water retention conservation, it is essential to first understand the dynamic changes in forests. In this study, the KO-G-Dynamic model, a Korean forest dynamic stand growth model, was used to analyze growth changes under different climate and forest management scenarios based on major tree species and forest types (Equation (1)). This model reflects the environmental and ecological characteristics of the Republic of Korea and can annually predict the growth of temperate forests based on stand volume dynamics. By incorporating climate change and forest management scenarios, it predicts future changes in forest resources and can support the development of optimal forest management plans [8,26,27,28].

$${\mathit{V}}_{ij}=a\times DB{H}_{ij}^b\times H{m}_{ij}^c\times N_{ij}$$

(1)

where $\mathit{V}$ is the stand volume, $\mathrm{i}$ represents the serial number for each stand, $\mathrm{j}$ denotes the year, $DBH$ is the stand average diameter at breast height, $Hm$ is the average tree height of the stand, and $N$ is the stand density (trees/ha). The coefficients (a, b, and c), determined for each tree species, were incorporated with the biomass allometric equation data developed by the NIFoS [29].

2.3. Forest Aboveground Water Yield Model

The aboveground water yield model was originally developed by Tian et al. (2008) [30] and later refined by Kim et al. (2021) [31] to reflect the characteristics of forests in the Republic of Korea, resulting in the development of the ‘Sanrim model.’ This model simulates the annual water yield by considering precipitation, species-specific evapotranspiration coefficients, slope, stand age, and surface runoff coefficients (Equation (2)) [30,31,32]. The developed model can be applied on species-specific coefficients to reflect the ecological characteristics of the Republic of Korea and is useful in predicting the water yield by incorporating forest management activities.

$$\begin{gathered} \mathit{M}{\mathit{D}}_{\mathit{i}}={\displaystyle \frac{1-E{T}_i}{1-E{T}_{CF}}}\times {\displaystyle \frac{C_i}{C_{25}}} \\ \mathit{W}{\mathit{S}}_{\mathit{t}}=\sum P\times \left(b_{ijt0}+b_{ijt1}\times P+b_{ijt2}\times P^2+\cdots \cdots +b_{ijtn}\times P^n\right)\times M{D}_i \end{gathered}$$

(2)

where $\mathit{M}{\mathit{D}}_{\mathit{i}}$ refers to the modifier at grid $\mathrm{i}$; $\mathit{E}{\mathit{T}}_{\mathit{C}\mathit{F}}$ is the evapotranspiration (ET) coefficient for Chinese fir, valued at 0.77, while $\mathit{E}{\mathit{T}}_{\mathit{K}}$ represents the $\mathrm{ET}$ coefficient for each of the major Korean tree species, red pine (0.70), Japanese larch (0.77), Korean pine (0.82), oak (0.60), and mixed forest (0.69), at grid $\mathrm{i}$. ${\mathit{C}}_{\mathit{i}}$ denotes the runoff coefficient based on slope at grid $\mathrm{i}$; ${\mathit{b}}_{\mathit{n}}$ are the coefficients for grid i, considering stand age $\mathrm{j}$ in year $\mathrm{t}$ [30]; $\mathrm{n}$ indicates the degree of the polynomial function; $\mathrm{P}$ refers to the precipitation at grid $\mathrm{i}$ during the spring season of year $\mathrm{t}$; and $\mathit{W}{\mathit{S}}_{\mathit{t}}$ represents the freshwater supply in year $\mathrm{t}$.

2.4. Forest Soil Belowground Water Storage Model

Many studies in the Republic of Korea have measured the mesopore ratio of parent rock and soil types distributed across the country and calculated the average soil depth by multiplying it by the area [33,34,35]. Based on these findings, the NIFoS developed an advanced estimation formula for the mesopore ratio in the A and B layers of broadleaf, conifer, and mixed forests. The formula utilizes soil mesopore ratio data derived from the site conditions of surveyed forest types and based on the increase in stand age (Equations (3)–(8)) [14]. Based on the results of stand age increase analyzed earlier, this study simulated belowground water storage.

$${\mathit{P}}_{\mathit{n}\mathit{a}}={\displaystyle \frac{40.0}{\left(1+\left(0.6e^{\left(-0.03AGE\right)}\right)\right)}}$$

(3)

$${\mathit{P}}_{\mathit{n}\mathit{b}}={\displaystyle \frac{35.0}{\left(1+\left(0.25e^{-0.01AGE)}\right)\right)}}$$

(4)

where ${\mathit{P}}_{\mathit{n}\mathit{a}}$ and ${\mathit{P}}_{\mathit{n}\mathit{b}}$ represent the mesopore ratio (%) of the A and B layers of coniferous forest soils, respectively, and $\mathrm{AGE}$ refers to the stand age in years.

For broadleaf forests, the mesopore ratios of the A and B layers are expressed as follows:

$${\mathit{D}}_{\mathit{n}\mathit{a}}={\displaystyle \frac{50.0}{\left(1+\left(0.29{\mathit{e}}^{\left(-0.01\mathit{A}\mathit{G}\mathit{E}\right)}\right)\right)}}$$

(5)

$${\mathit{D}}_{\mathit{n}\mathit{b}}={\displaystyle \frac{45.0}{\left(1+\left(0.6{\mathit{e}}^{\left(-0.02\mathit{A}\mathit{G}\mathit{E}\right)}\right)\right)}}$$

(6)

where ${\mathit{D}}_{\mathit{n}\mathit{a}}$ and ${\mathit{D}}_{\mathit{n}\mathit{b}}$ represent the mesopore ratio (%) of the A and B layers of broadleaf forest soils, respectively, and $\mathrm{AGE}$ refers to the stand age in years.

Similarly, for mixed forests, the mesopore ratios of the A and B layers are represented as follows:

$${\mathit{M}}_{\mathit{n}\mathit{a}}={\displaystyle \frac{45.0}{\left(1+\left(0.6{\mathit{e}}^{\left(-0.02\mathit{A}\mathit{G}\mathit{E}\right)}\right)\right)}}$$

(7)

$${\mathit{M}}_{\mathit{n}\mathit{b}}={\displaystyle \frac{40.0}{\left(1.2+\left(4.0{\mathit{e}}^{\left(-0.07\mathit{A}\mathit{G}\mathit{E}\right)}\right)\right)}}$$

(8)

where ${\mathit{M}}_{\mathit{n}\mathit{a}}$ and ${\mathit{M}}_{\mathit{n}\mathit{b}}$ represent the mesopore ratio (%) of the A and B layers of mixed forest soils, respectively, and $\mathrm{AGE}$ refers to the stand age in years.

2.5. Data Preparation and Modification

In this study, baseline data were established to analyze water retention conservation under forest management strategies for climate change adaptation (Figure 2).

To quantitatively assess the impacts of climate change, climate data from the Korea Meteorological Administration (KMA) were utilized. The historical climate data consist of automated synoptic observing system (ASOS) data collected from 95 observation stations between 2011 and 2020. For future climate change data, the average temperature and precipitation derived from the 5ENSMN models (GRIMs, RegCM, WRF, CCLM, HadGEM3-RA), based on MK-PRISM, were applied. These data, representing the detailed SSP5-8.5 scenario for the Republic of Korea, were applied at a 1 km × 1 km resolution for the period of 2021–2100 (Figure 3) (see Supplementary Materials) [36].

To simulate forest changes, data from the 1:5,000 forest type map and the 6th NFI, including forest type, species, age class, site index, DBH, and canopy density, were used (see Supplementary Materials) [37,38]. Slope information was derived from Digital Elevation Model (DEM) data provided by the National Geographic Information Institute (NGII) [28]. Soil characteristics, including soil type distribution and effective soil depth, were obtained from the national soil map for model execution (see Supplementary Materials) [39]. To simulate water retention conservation under different forest management conditions, the normal and ideal forest management scenarios were established and formally defined in this study. These strategies were designed to reflect current operational practices and policy-optimized approaches, respectively. For forest management scenarios, thinning and clear-cutting areas were determined based on stand volume and domestic wood supply data from the 6th National Forest Plan and 4th Five-Year Forest Tending Promotion Plan. For the current forest management scenario, the ‘Normal Forest Management’ involves approximately 120,000 ha of forest trending annually, while the ‘Ideal Forest Management’ scenario, based on optimized forest planning, assumes approximately 200,000 ha of forest tending [27]. In this scenario, after clear-cut harvesting and thinning, considering the legal cut and normal final age, the areas where clear-cut harvesting has occurred are assumed to be replanted with climate change-appropriate species suitable for the region [28,40,41].

For the watershed information, the catchment area watershed map (average area of about 267 ha), which can encompass the detailed stand-level units of forests (average area of about 322 ha) from the watershed division map in the Water Environment Information System, was applied and analyzed in this study (see Supplementary Materials) [22].

3. Results

Based on the simulation approach described above, the total annual water resources in the Republic of Korea were estimated to be approximately 36.9 billion tons in 2023, showing a similar trend to previous studies that reported 36.6 billion tons [10].

Furthermore, the analysis indicated that if an ideal forest management approach, based on the 6th National Forest Plan and the 4th Five-Year Forest Tending Plan, aligned with climate change adaptation policies, were implemented instead of the current normal forest management practices, additional water retention of 0.309 billion tons could be secured in the 2030s (2026–2035), 0.702 billion tons in the 2050s (2046–2055), and 1.188 billion tons in the 2080s (2076–2085) (Figure 4).

3.1. Forest Aboveground Water Yield

In regard to the aboveground water yield, under the normal forest management scenario, if the current trend continues, it is predicted that 2.612 billion tons per year will be yielded in the 2030s, 2.312 billion tons per year in the 2050s, and 3.427 billion tons per year in the 2080s. However, if a watershed-based ideal forest management approach is implemented as per forest policy, it is predicted that 2.658 billion tons per year will be supplied in the 2030s, 2.400 billion tons per year in the 2050s, and 3.609 billion tons per year in the 2080s (Figure 5).

The analysis showed that the aboveground water yield is expected to decline as climate change progresses. However, depending on the level of forest management, the supply trend may improve in the medium to long term. This is attributed to the fact that conifer species, which dominate most forests in the Republic of Korea, exhibit higher evapotranspiration rates than broadleaf species.

Additionally, as climate change advances, forest zones are shifting northward, leading to an increasing dominance of broadleaf trees over conifers. Consequently, evapotranspiration is expected to decrease. However, due to repeated mortality and succession, a trend similar to that illustrated in Figure 6 is predicted.

3.2. Forest Soil Belowground Water Storage

The belowground water storage results exhibited a trend similar to the aboveground component (Figure 7). This can be attributed to increased water absorption by forest stands and enhanced water storage in soil layers, which occurred when ideal watershed-based forest management was implemented. This approach, based on forest policy, alleviated the imbalance in age classes compared to current forest management trends.

Under the continuation of the normal forest management scenario, it is predicted that for the A layer of soil, 16.645 billion tons per year will be stored in the 2030s, 17.103 billion tons per year in the 2050s, and 17.523 billion tons per year in the 2080s. For the B layer of soil, estimates indicate storage of 18.296 billion tons per year in the 2030s, 18.859 billion tons per year in the 2050s, and 19.356 billion tons per year in the 2080s.

If ideal forest management is implemented, it is predicted that for the A layer of soil, 16.791 billion tons per year will be stored in the 2030s, 17.325 billion tons per year in the 2050s, and 18.052 billion tons per year in the 2080s. For the B layer of soil, analysis indicates that 18.414 billion tons per year will be stored in the 2030s, 19.152 billion tons per year in the 2050s, and 19.834 billion tons per year in the 2080s (Figure 8).

4. Discussion

Currently, coniferous species in the forests of the Republic of Korea cover approximately 1.2 times the area and account for about 1.7 times the growing stock of broadleaf species [23]. The evapotranspiration rate of coniferous species, which dominate most forests, is higher than that of broadleaf species [42]. However, as climate change progresses, forest zones are shifting northward, leading to an increasing dominance of broadleaf species over conifers, which is expected to result in decreased evapotranspiration. Despite this, conifers remain highly vulnerable to climate change due to their high mortality rates, and repeated succession and disturbances make it challenging to secure water resources within forests. This study aims to contribute to the development of watershed-based forest management plans by exploring strategies to enhance forest water retention conservation in response to climate change.

According to the NIFoS, the long-term water resource outlook for the Republic of Korea predicts that water shortages and droughts will intensify due to decreasing total runoff and seasonal discharge until the 2040s [43]. Furthermore, it has been suggested that to offset the drought projections of the 2000s, between 121,000 and 393,000 hectares of forest tending would be required annually [10]. Previous studies have also predicted water resource-securing strategies through watershed-level and optimal management strategies [31,44]. In that regard, the findings of this study align with these projections, demonstrating that water retention conservation increases with the implementation of sustainable forest management. However, a more detailed examination of the aboveground water yield showed that while forest management led to a slight decrease in the medium to long term due to the impact of stand growth and biomass, in the long term, water resource availability was facilitated by the alleviation of age-class imbalance and the subsequent increase in growth and stocking.

For belowground water storage, the analysis indicated a higher water retention conservation capacity and a greater annual increase compared to the aboveground water yield. This is due to the well-differentiated soil layers with distinct physical properties, ranging from the organic layer to the B layer. The A layer contains numerous pores that facilitate water storage, as the abundant organic matter causes inorganic particles to aggregate into structures of various sizes. Additionally, the A layer features pathways created by microfauna that feed on organic matter and vessels formed by decaying roots, creating a network that allows for the rapid absorption of large amounts of rainfall. The B layer, in turn, stores the water absorbed by the A layer at greater depths and releases it gradually. Although the B layer has a lower pore ratio due to its lower organic matter content, its greater depth enables it to store a larger volume of rainfall [14,45,46].

Moreover, this study suggests that replanting climate change-appropriate species in clear-cut harvested areas following forest tending contributed to a decrease in evapotranspiration. This improvement in age-class structure and the increasing proportion of broadleaf species resulted in enhanced water retention conservation. Additionally, regions such as Gangwon Province, Gyeongsangnam-do, and Gyeongsangbuk-do, which have higher growing stock than the national average, demonstrated a tendency for greater water retention conservation. A correlation was also observed between age-class structure and the relationship between water supply and retention conservation.

The models used in this study have been widely applied in both domestic and international research and have been published in the academic literature [14,31,40]. Furthermore, the findings of this study were validated by comparison with previous research, reinforcing their significance [43]. This study provides valuable insights by classifying and predicting forest water resource dynamics based on watershed units in response to climate change, offering a pathway for sustainable forest management. Furthermore, it departs from conventional forest management systems based on legally defined administrative boundaries and instead adopts ecologically relevant watershed units, thereby enhancing spatial applicability and reflecting environmental characteristics more accurately. However, a key limitation is that the study does not fully account for the complex physiological mechanisms of forests, the impact of disturbances, and climate variability. Additionally, it does not reflect the spatial heterogeneity of soil properties below the B layer and the dynamics of deep groundwater in soil water storage calculations. Future research should focus on process-based water retention conservation predictions to address these gaps [47,48,49].

Since the era of erosion control and greening, the Republic of Korea has consistently worked to secure forest water resources. Recently, the increasing importance of forest water resources, driven by worsening water shortages, has been emphasized in the 6th National Forest Plan. In response to the declining supply of forest water resources, there is a growing focus on expanding water retention conservation through forest tending and strengthening the management of designated water retention conservation zones. As a result, a watershed-based forest management system has been established, with plans formulated to secure forest water resources [25]. This study not only provides foundational data for national and regional forest planning by spatially predicting water resources in watershed-based standard forest units in response to climate change but also contributes scientifically to climate adaptation policies.

5. Conclusions

This study aimed to analyze water retention conservation under climate change and different forest management scenarios, proposing strategies to secure sustainable water resources through watershed-based forest management. To achieve this, the KO-G-Dynamic model—a forest dynamic stand growth model—along with aboveground and belowground water resource prediction models, was applied to evaluate water retention conservation under climate change scenarios and forest management strategies. The results were validated through comparisons with previous studies.

If climate change continues, water supply in the Republic of Korea is projected to decline by 5.9% in 2030 and by 9.4% in 2037 [11]. However, if the current forest management practices persist, water supply is expected to increase by 1.77% in the 2030s, 1.92% in the 2050s, and 5.31% in the 2080s. In contrast, if ideal forest management is implemented in alignment with forest policies, water supply is projected to increase by 2.61% in the 2030s, 2.68% in the 2050s, and 6.73% in the 2080s. These findings suggest that a watershed-based forest management system is not only feasible but also crucial for ensuring water resource availability under climate change.

This study highlights that science-based predictions of forest water retention conservation can support policy development by facilitating the transition from a uniform forest water management approach—outlined in the 6th National Forest Plan—to a watershed-based system. Furthermore, the generation of spatial and temporal maps of forest water retention conservation can serve as valuable references for regional forest management planning.