Impacts of Climate and Land Use/Land Cover Change on Water Yield Services in Heilongjiang Province

Abstract

The Heilongjiang Province is the region in China with the highest grain production and the most significant ecological security barrier in the northern part of the country. In recent years, the growing necessity for water in human production and life has intensified water resource conflicts. In order to identify a solution to this situation, the InVEST model’s water yield module is employed to simulate the water yield in Heilongjiang Province in 2000, 2010, and 2020. The water yield in Heilongjiang Province from 2000 to 2020 is evaluated, and the scenario simulation method is employed to investigate the impact of climate change and land use type changes on water yield. The results indicate that from 2000 to 2020, the annual precipitation in Heilongjiang Province increased from 490 mm in 2000 to 764 mm in 2020. The spatial distribution shows a west–high and east–low pattern, with Jixi City having the highest average annual precipitation and the Greater Khingan region having the lowest. During the period of 2000–2020, woodland had the highest proportion, followed by arable land, while built-up land had the smallest proportion. The conversion of unused land and woodland represents the primary driver of the expansion in farmland areas, while the conversion of water bodies accounts for the majority of the growth in the area of unused land. The areas of woodland and water bodies exhibited a slight decrease. The order of water yield of land use types, from the greatest to the least, is as follows: built-up land, unused land, farmland, grassland, woodland, and water bodies. The main factor affecting changes in precipitation levels in the province is climate change, which contributed up to 99.58% during the period 2000–2020. In contrast, changes in land use types contributed a mere 0.42%. The sustained expansion of the urban population in Heilongjiang province has resulted in an augmented water yield in select regions.

1. Introduction

The ecosystem represents the fundamental unit of Earth’s life support system. The stability of this system is inextricably linked to the survival and development of humanity. It performs several functions, including the purification of water quality, the maintenance of soil and water, the cleaning of air, and the preservation of biological diversity [1]. The water yield plays an integral role in the sustainable development of regional economies and ecosystems, as it constitutes a fundamental component of the ecosystem service function [2,3,4]. It is commonly accepted that the water yield service is essentially equivalent to the water yield of the ecosystem. Both are calculated to subtract actual evapotranspiration from precipitation [5]. We are currently facing unprecedented global climate change, and the concomitant increase in human demand for water resources has led to a situation of water scarcity, which has severely impacted the sustainability of regions [6,7,8].

The spatiotemporal variation in ecosystem water yield services is significantly impacted by climate change and land use/land cover change [9,10,11]. While the process of climate change is gradual, its consequences are substantial [12]. Climate change exerts its primary influence on water yield through alterations in precipitation and evapotranspiration [13]. PESSACG et al. [14] demonstrated that alterations in precipitation levels can result in notable variations in water yield. Jorda-Capdevila D et al. [15] demonstrated that a reduction in precipitation levels is associated with a notable decline in water yield. The observed changes in land use are primarily driven by ecological policies, with relatively minor but rapid impacts. The most typical example is the development of urbanization and the expansion of arable land, which has resulted in a decrease and degradation of other types of land [16,17]. The alteration of land use directly affects the underlying surface conditions, thereby influencing evapotranspiration and infiltration processes and, consequently, the quantity of water yielded [18]. Gao et al. [19] analyzed the relationship between land use and water production. The findings indicated that as urbanization increases, there is a corresponding rise in impervious surfaces and water yield. Zhang Wei et al. [20] investigated the response of water yield in the Zhang Heng watershed to land use changes. The results of these studies collectively demonstrate that both climate change and land use/land cover change are exerting a considerable influence on the water yield of ecosystems. However, most of these studies focus on the effects of individual factors, with few studies providing comprehensive assessments and quantification of the combined effects of these two factors on water yield.

The InVEST model provides a simple quantitative method for calculating water yield under different conditions [21,22], with flexible parameter settings and wide applicability. It has been employed extensively in research and has yielded favorable outcomes across a spectrum of disciplines. Dou Panfeng, Zuo Shuzhai, Ren Yin, et al. [23] examined the effects of climate and land use changes on water services in the Ningbo area. Furthermore, scenario analysis was employed to ascertain the contribution rate of climate and land use on alterations in water yield. Zhao Yaru, Zhou Junju, Lei Li, et al. [24] analyzed the process of change in water yield in the upper reaches of the Shiyang River Basin. In addition, the driving factors for water yield were further investigated using DeoDetectors. Wu Jian, Li Yinghua, Huang Liya, et al. [25] used the InVEST model to study the distribution of water yield in the Northeast region from 1990 to 2010. They used structural equation modeling to investigate the effects of climate factors and land use change on water yield in the Northeast region. Lang Y et al. [2] found that alterations in precipitation levels exert a considerable influence on water yield. In summary, the In-VEST model can provide a visual assessment of the spatio-temporal evolution of regional water yield at different scales, times, and factors. This allows evaluation in different areas. Currently, there is limited research on water yield in Heilongjiang Province. As the northernmost province in China, Heilongjiang’s geographical location and ecological environment are of great importance. It is, therefore, of great importance to study the water yield in Heilongjiang Province.

The Heilongjiang Province has consistently ranked first in the country in grain production for 14 consecutive years and is also the most typical cold region in Northeast China. It bears a heavy responsibility for national food security and environmental safety, and its water yield directly affects crop yields and environmental safety. However, the existing body of research on water yield services in Heilongjiang Province is inadequate, and the factors influencing the functionality of these services in the province remain uncertain. Based on this, this study aims to investigate the spatiotemporal variation characteristics and influencing factors of water yield in Heilongjiang Province. The aim is to provide scientific evidence for water resource planning, food production safety, and environmental safety in Heilongjiang Province.

2. Materials Data and Methods

2.1. Overview of the Research Area

The Chinese province of Heilongjiang is situated in the country’s northeastern region. It is the northernmost and easternmost province in China, as well as the province with the highest latitude, encompassing a range of longitudes between 121°11′ E and 135°05′ E and latitudes between 43°26′ N and 53°33′ N (Figure 1). It is bordered to the west by the Inner Mongolia Autonomous Region and to the south by Jilin Province. It faces Russia across the river to the north and east. The Heilongjiang Province exhibits a temperate continental monsoon climate, with temperatures and precipitation levels typically low during the spring season. Summer months are marked by warm and humid conditions, while the autumn season is prone to flooding and early frosts. Winters are long and cold, with significant snowfall. Furthermore, Heilongjiang Province occupies the fourth position in the country with a forest area. The province of Heilongjiang is distinguished by a profusion of rivers and lakes, which are dispersed across the four principal water systems: the Heilong River, Ussuri River, Songhua River, and Suifen River. A total of 2881 rivers with a basin area of 50 square kilometers or more have been identified, collectively covering a length of 92,100 km. The lakes in Heilongjiang Province include Khanka Lake, Jingpo Lake, and Wudalianchi Lake. A total of 253 lakes resulting in a water surface area of 1 square kilometer or more were identified. The topography is elevated in the northwestern, northern, and southeastern regions, while the northeastern and southwestern regions are relatively low in elevation. The region is mainly characterized by mountainous terrain, platforms, plains, and bodies of water. The majority of the mountainous areas in Heilongjiang Province exhibit elevations ranging from 300 to 1000 m. The elevation of a platform is between 200 and 350 m, while that of a plain is between 50 and 200 m. The urbanization rate of Heilongjiang Province has consistently exceeded the national average for China. However, in recent years, the population outflow has resulted in a decline in the number of permanent residents.

2.2. Research Methodology

2.2.1. Water Yield Model

The InVEST model is based on the Budyko curve and the mean annual precipitation. The annual water yield, designated as Y(x), is calculated for each pixel within the grid [18]:

$$\mathrm{Y}(\mathrm{x})=1-{\displaystyle \frac{\mathrm{AET}(\mathrm{x})}{\mathrm{P}(\mathrm{x})}}·\mathrm{P}\left(\mathrm{x}\right)$$

where AET(x) represents the grid pixel, x is the actual annual evapotranspiration, and P(x) is the precipitation at pixel x.

The water balance equation adopts the Budyko hypothesis on the coupled water–energy balance assumption formula [26].

$${\displaystyle \frac{\mathrm{AET}\left(\mathrm{x}\right)}{\mathrm{P}\left(\mathrm{x}\right)}}=1+{\displaystyle \frac{\mathrm{PET}\left(\mathrm{x}\right)}{\mathrm{P}\left(\mathrm{x}\right)}}-{\left[1+{({\displaystyle \frac{\mathrm{PET}\left(\mathrm{x}\right)}{\mathrm{P}\left(\mathrm{x}\right)}})}^{\mathsf{\omega }\left(\mathrm{x}\right)}\right]}^{{\displaystyle \frac{1}{\mathsf{\omega }\left(\mathrm{x}\right)}}}$$

where PET(x) represents potential evapotranspiration, and ω(x) is a non-physical parameter characterizing natural climate–soil properties.

Potential evapotranspiration (PET) is defined as follows:

$$\mathrm{PET}(\mathrm{x})={\mathrm{K}}_{\mathrm{c}}\left({\mathrm{l}}_{\mathrm{x}}\right)\times {\mathrm{ET}}_0\left(\mathrm{x}\right)$$

where ET₀(x) is the potential evapotranspiration of pixel x, and K_c(lx) is the evapotranspiration coefficient for the land use type at pixel x. ω(x) represents the empirical parameter, and the InVEST model adopts the calculation formula proposed by Donohue et al. [27]:

$$\mathsf{\omega }(\mathrm{x})=\mathrm{Z}{\displaystyle \frac{\mathrm{AWC}\left(\mathrm{x}\right)}{\mathrm{P}\left(\mathrm{x}\right)}}+1.25$$

The AWC(x) represents the available soil water content of grid unit x; Z is an empirical constant known as the seasonal factor.

$$\mathrm{AWC}(\mathrm{x})=\mathrm{Min}\left(\mathrm{Rest}.\mathrm{layer}.\mathrm{depth},\mathrm{root}.\mathrm{depth}\right)·\mathrm{PAWC}$$

where PAWC is used to denote the plant’s available water content, which is calculated using the empirical formula proposed by Zhou et al. [28]:

$$\begin{gathered} \mathrm{PAWC}=54.509-0.132\mathrm{SAND}-0.003{(\mathrm{SAND})}^2-0.055\mathrm{SILT}-0.006{(\mathrm{SILT})}^2-0.738\mathrm{CLAY} \\ +0.007{(\mathrm{CLAY})}^2-2.699\mathrm{OM}+0.501{\left((\mathrm{OM})\right)}^2 \end{gathered}$$

where SAND, SILT, CLAY, and OM are the sand, clay, silt, and organic matter contents, respectively.

2.2.2. Climate and Land Use Scenarios

In order to ascertain the water yield in Heilongjiang Province in 2000, 2010, and 2020, the InVEST model was employed for simulation purposes. In order to explore the impact of climate change and land use/land cover change on water yield, the study used a scenario analysis method. Firstly, the design scenario assumes that the land use types will remain unaltered while the climate undergoes a change. Secondly, the design scenario assumes that the climate will remain unaltered while land use types undergo a change. By comparing with the actual scenario, the effects of climate change and changes in land use/land cover type on water yield in Heilongjiang Province can be studied separately in three different time periods (

Table 1. Scenarios of climate change and land use/land cover change.

	Actual Scenarios			Climate Change Scenarios			Land Use Type Changes Scenarios
Scenario	2000	2010	2020	Scenario 1	Scenario 2	Scenario 3	Scenario 4	Scenario 5	Scenario 6
Climate Elements	2000	2010	2020	2020	2010	2020	2000	2000	2010
Land use types	2000	2010	2020	2000	2000	2010	2020	2010	2020

).

The impact of climate change and land use type change on water yield can be quantified by examining the changes in water yield under different scenarios using the following formula:

$$C_c={\displaystyle \frac{c}{c+s}}\times 100\%$$

$$C_s={\displaystyle \frac{s}{c+s}}\times 100\%$$

where the impact of climate change on water yield is expressed as C_c, while the impact of land use type change on water yield is expressed as C_s. The variables C_c and C_s represent the changes in water yield under the climate change and land use type change scenarios, respectively.

2.3. Data Sources

(1) Precipitation, temperature, evaporation, wind speed, and sunshine hours are obtained from 34 meteorological stations in Heilongjiang Province. Potential evaporation is calculated using the FAO Penman–Monteith formula [29], and meteorological grid data are obtained by inverse distance-weighted interpolation.

(2) Land use/land cover data are from the China land use/cover change dataset provided by the Resources and Environmental Science Data Platform [30].

(3) Soil data are from the Harmonised World Soil Database, produced by the FAO [31].

(4) The depth of the root restriction layer is selected from the Depth-to-Bedrock map of China [32].

(5) The biophysical parameter table is determined on the basis of the In-VEST model manual, “Crop Evapotranspiration—Guidelines for Computing Crop Water Requirements—FAO Irrigation and drainage paper 56” and relevant research [29].

(6) The data regarding population are derived from the Heilongjiang Province Statistical Yearbook.

2.4. Model Validation

The simulated water yields for the years 2000, 2010, and 2020 are 475.26 billion m³, 735.13 billion m³, and 1218.76 billion m³, respectively. These figures were compared with those published in the Heilongjiang Statistical Yearbook, which reported water yield volumes of 479.22 billion m³, 725.3 billion m³, and 1221.5 billion m³ for the same years. The relative errors for all comparisons were within 2%, indicating that the simulated water yield results are relatively accurate and reliable when compared to the official statistics from the Yearbook.

3. Results and Analysis

3.1. Changes in Climatic Factors

Precipitation in Heilongjiang Province was 490 mm, 540 mm, and 765 mm in 2000, 2010, and 2020, respectively (Figure 2). Compared to the year 2000, precipitation increased by 275 mm (56.12%) in 2020 and by 50 mm (10.2%) in 2010. Compared to 2010, precipitation in 2020 increased by 225 mm (41.67%). It is noteworthy that precipitation levels have increased significantly between 2000 and 2020, by up to 56.12%. This can be attributed to the exceptional nationwide drought that China experienced in 2000, which was one of the most severe drought years since the founding of the country, and the northern region of China experienced the most severe drought since 1980. In contrast, the precipitation levels in Heilongjiang Province in 2020 were the highest recorded since 1961. In 2000, 2010, and 2020, the potential evapotranspiration in Heilongjiang Province was 804 mm, 736 mm, and 814 mm, respectively (Figure 2). Compared to the year 2000, there is an increase of 10 mm (1.24%) in potential evapotranspiration in 2020, while there is a decrease of 68 mm (8.46%) in 2010. Compared to the year 2010, there is an increase of 78 mm (10.6%) in potential evapotranspiration in 2020. In general, areas with higher precipitation and lower potential evaporation tend to have a greater water yield, while those with less precipitation and higher potential evaporation tend to have a smaller water yield.

3.2. Land Use/Land Cover Change

In 2020, 44.75% of the total area of Heilongjiang was comprised of woodlands, 35.38% was devoted to agriculture, 7.17% was designated for grassland, and 7.79% was classified as unused land. Additionally, 2.09% of the total area was occupied by water bodies, while 2.1% was designated for built-up land. During the period 2000–2020, changes in land use/land cover occurred (Figure 3,

Table 2. Transition matrix of land use types/land cover unit: km².

		Grassland	Built-Up Land	Farmland	Woodland	Waterbody	Unused	Total
	2020
2000	Grassland	27,524	132	1546	1075	184	1516	31,977
	Built-up land	27	8037	616	46	23	25	8774
	Farmland	781	1092	155,125	1488	331	1265	160,082
	Woodland	3235	131	2835	199,238	746	1480	207,665
	Waterbody	211	43	521	324	7909	5750	14,758
	Unused	682	82	2680	301	241	25,189	29,175
	Total	32,460	9517	163,323	202,472	9434	35,225	452,431
	2010
2000	Grassland	12,708	194	4158	7335	504	7080	31,979
	Built-up land	145	6130	2223	130	58	88	8774
	Farmland	5266	2911	140,117	6124	1623	4042	160,083
	Woodland	12,543	232	9477	178,237	861	6315	207,665
	Waterbody	1212	84	1002	362	8913	3185	14,758
	Unused	3722	120	5859	2392	1609	15,475	29,177
	Total	35,596	9671	162,836	194,580	13,568	36,185	452,436
	2020
2010	Grassland	14,058	211	5527	10,866	519	4414	35,595
	Built-up land	163	6221	2876	198	63	148	9669
	Farmland	3469	2674	142,646	8439	463	5146	162,837
	Woodland	7997	182	6207	176,787	582	2824	194,579
	Waterbody	323	81	1303	524	6720	4618	13,569
	Unused	6450	147	4765	5658	1089	18,075	36,184
	Total	32,460	9516	163,324	202,472	9436	35,225	452,433

). During the period from 2000 to 2020, the grassland area increased from 31,978 km² in 2000 to 35,595 km² in 2010 and then decreased to 32,459 km² in 2020, with a total increase of 481 km². This increase is mainly due to the conversion from woodland (3235 km²). The built-up land area increased from 8774 km² in 2000 to 9670 km² in 2010 and then decreased to 9517 km² in 2020, with an increase of 743 km², mainly converted from farmland (1092 km²). The farmland area increased from 160,081 km² in 2000 to 162,836 km² in 2010 and then decreased to 163,324 km² in 2020, with an increase of 3243 km². This increase was mainly due to the conversion of woodland (2835 km²), unused land (2680 km²), and grassland (1546 km²). The woodland area decreased from 207,665 km² in 2000 to 194,579 km² in 2010 and then increased to 202,472 km² in 2020; the total area decreased by 5193 km², mainly converted into grassland (3235 km²), farmland (2835 km²), and unused land (1480 km²). The waterbody decreased from 14,757 km² in 2000 to 13,569 km² in 2010 and further to 9434 km² in 2020, with a total decrease of 5323 km², the main conversion was into unused land (5750 km²). The unused area increased from 29,176 km² in 2000 to 36,184 km² in 2010 and then decreased to 35,225 km² in 2020, with a net increase of 6049 km², mainly due to the conversion of waterbody (5750 km²).

3.3. Changes in the Permanent Population and Urbanization Rate

According to the “Heilongjiang Statistical Yearbook”, the permanent population of Heilongjiang in 2000, 2010, and 2020 was 21.71 million, 38.33 million, and 38.07 million (Figure 4), respectively, with urbanization rates of 51.94%, 55.66%, and 65.6%; the urban population was 16.47 million, 21.33 million, and 24.97 million, respectively.

3.4. Water Yield Changes

The data regarding water production in Heilongjiang Province were obtained through the use of a simulation calculation. The water yield in the years 2000, 2010, and 2020 were 105 mm, 162 mm, and 269 mm, respectively. In comparison to the yield observed in 2000, there was an increase of 57 mm (54.28%) in 2010 and a further increase of 164 mm (156.19%) in 2020. Furthermore, the water yield in 2020 exhibited an increase of 107 mm (66.05%) in comparison to 2010. The interannual variation in the water yield during the period from 2000 to 2010 was 5.7 mm, which is lower than that observed during the period from 2010 to 2020 when the interannual variation was 10.7 mm.

As illustrated in Figure 5, alterations in land use/land cover have a discernible effect on water yield. In 2000, the water yield of land use types ranked from highest to lowest as follows: built-up land (274 mm), unused land (233 mm), farmland (135 mm), grassland (104 mm), woodland (62 mm), waterbody (2 mm). In 2010, the order of water yield of land use types, from the greatest to the least, is as follows: built-up land (348 mm), unused land (313 mm), farmland (194 mm), grassland (153 mm), woodland (86 mm), waterbody (5 mm). In 2020, the order of water yield of land use types, from the greatest to the least, is as follows: built-up land (568 mm), unused (471 mm), farmland (160 mm), grassland (269 mm), woodland (160 mm), waterbody (19 mm). A consistent upward trajectory was observed in the water yield from all land use/land cover types between the years 2000 and 2020.

From a spatial perspective, the high-yield water areas in Heilongjiang Province are mainly located in the central and eastern regions, while the low-yield water areas are mainly located in the northwest (Figure 6). From 2000 to 2020, the water yield in Mudanjiang City and Qitaihe City initially exhibited a decline, followed by an increase. In contrast, the water volume in other regions demonstrated a consistent upward trend. In 2000, the water yield in Jixi City was the highest, while it was the lowest in the Greater Khingan region. In 2010, Jiamusi township had the highest average water yield, while Greater Khingan region township had the lowest. In 2020, Jixi City exhibited the highest water yield in the Heilongjiang Province, while the Greater Khingan region demonstrated the lowest water yield (Figure 7).

3.5. Impact of Climate Change on Water Yield

Climate change directly causes changes in precipitation and potential evapotranspiration, thereby affecting water yield under scenarios 1, 2, and 3 without land type change. (Figure 8 and Figure 9). The water yield of scenario 1 is 312 mm, which is an increase of 207 mm (197.14%) compared to the actual scenario in 2000. The quantity of water yielded in all regions has increased, with high-yield areas, including Suihua, Jiamusi, Qitaihe, and Jixi, and low-yield areas, including the Greater Khingan Range region. The water yield of Scenario 2 is 150 mm, representing an increase of 45 mm (42.86%) compared to the actual situation in 2000. The water production in Mudanjiang and Qitaihe has decreased, whereas it has increased in other areas. The regions with the highest water production are Qiqihar, Qitaihe, and Jixi, while the region with the lowest water production is the Great Khingan region. The water yield of Scenario 3 is 328 mm, which has increased by 166 mm (102.47%) compared to the actual situation in 2010. The quantity of water yielded in all regions has increased, with higher yields in Suihua City, Jiamusi City, Qitaihe City, and Jixi City, while the lower-yielding area is the Great Khingan region. Under normal circumstances, water yield will increase as precipitation increases and decrease as evaporation increases.

3.6. Impact of Land Use/Land Cover Change on Water Yield

Transformations in land use and land cover types have the potential to result in alterations to the local evapotranspiration capacity, surface water infiltration capacity, and water retention capacity, thereby affecting overall water yield. Scenarios 4, 5, and 6 represent situations where climate remains unchanged (Figure 10 and Figure 11). The water yield of Scenario 4 is 106 mm, which is 1 mm (0.95%) higher than the actual situation in 2000. With the exception of Hegang, Jiamusi, and Shuangyashan, the water yield in other regions has increased. The distribution trend of water yield is quite similar to the actual situation in 2000. The water yield of scenario 5 is 108 mm, which has increased by 3 mm (2.86%) compared with the actual situation in 2000. The water yield has decreased in the cities of Heihe, Jixi, Jiamusi, Qitaihe, and Suihua, while it has increased in other areas. The high-yield areas are concentrated in Harbin, Mudanjiang, and Jixi, while the low-yield areas show a distribution trend like that of 2000. The water yield of scenario 6 is 119 mm, which is 43 mm (26.54%) lower than the actual situation in 2010. Water yield has decreased in all regions, with high water yield areas in Jiamusi and Jixi cities and low water yield areas in the Greater Khingan region.

3.7. The Relationship between Water Yield, Land Use/Land Cover Type, and Climate

It can be reasonably inferred that climate factors exert a more significant influence on water yield in the three time periods under consideration: 2000–2010, 2010–2020, and 2000–2020 (

Table 3. Contribution rate of climate change/land use type change to water yield.

Year	Factors of Influence	Contribution Rate
2000–2010	climate change	93.75%
	land use type change	6.25%
2010–2020	climate change	79.43%
	land use type change	20.57%
2000–2020	climate change	99.52%
	land use type change	0.48%

).

4. Discussion

4.1. Impact of Land Use/Cover on Water Yield Services

Research has shown that built-up land use results in the highest water yield. As homes and roads are built, materials such as concrete and asphalt prevent rainwater from soaking into the ground, leading to increased surface runoff. In addition, the transpiration effect of vegetation reduces water yield, and since built-up land has less vegetation cover, it results in higher water yield compared to other land uses [2]. However, in built-up land, precipitation typically flows directly into drainage systems and is difficult to use [33]. Accordingly, further study is required to identify methods for enhancing the utilization rate of precipitation on built-up land.

The lowest water yield is woodland due to interception of precipitation by vegetation, absorption of precipitation by surface litter and roots, storage and infiltration of soil layers, and transpiration of vegetation during the precipitation process. All of these factors contribute to a reduction in surface runoff [34,35]. Grassland, farmland, and forest ecosystems are similar, with low water yield. However, due to the higher plant density and shallower root system of farmland and grassland, their capacity to regulate precipitation is inferior to that of woodland [36]. Therefore, woodlands have lower water yields than these ecosystems. In contrast to the situation in woodlands, the introduction of wetland conservation measures has resulted in a net increase in the extent of unused land, accompanied by a corresponding rise in water volume. In formulating pertinent policies, it is imperative to consider the impact of land use/land cover changes on water yield.

4.2. Impact of Climate Change on Water Yield

The study demonstrated that climate exerts a more pronounced influence on water yield when only climate and land use/land cover factors are taken into account. The contribution rates of climate change to water yield in Heilongjiang Province in the three time periods of 2000–2010, 2010–2020, and 2000–2020 are 93.75%, 79.43%, and 99.52%, respectively, while the contribution rates of land type change to water yield in Heilongjiang Province for the same time periods are only 6.25%, 20.57%, and 0.48%. This is because the change process and influence mechanism are relatively complex. The conversion of land use types may result in alterations to water yield; however, the overall change in water yield is not substantial [37].

Precipitation and evapotranspiration are two important climatic variables [26]. The spatial distribution of water yield exhibits a correlation with that of precipitation. The quantity of precipitation exerts a direct influence on human production and life, as well as on water yield services. An excess of precipitation may precipitate a flood disaster, whereas an insufficient quantity may result in a situation of water resource scarcity. The quantity of precipitation exerts a direct influence on agricultural production and ecological security [38]. Evapotranspiration exerts a comparable influence on the provision of water yield services. A proportion of the available water resources is lost to the atmosphere as a result of evapotranspiration prior to utilization by humans. This has the effect of reducing both the utilization rate and the quality of the remaining water resource [39].

4.3. The Impact of Population and Urbanization Rate on Water Yield

Due to the depletion of resources in some cities within Heilongjiang Province, the population dividend relying on resource development has been exhausted, and the region is experiencing the pains of industrial structural transformation, leading to population outflow. Since 2010, the resident population of Heilongjiang Province has slightly decreased, but the implementation of urbanization policies has led to a continuous increase in the urbanization rate. This has resulted in sustained growth of urban population and expansion of construction land, leading to increased water consumption in certain areas.

4.4. Limitations and Uncertainties

The selection of empirical formulas for calculation purposes may influence the outcome of the simulation, as some of the data in the model are calculated based on these formulas. It is possible that some errors may be generated. Therefore, it is necessary to determine the most appropriate empirical formulas and parameters according to local conditions to improve the accuracy of simulation results. Furthermore, this study covers a wide range of years, but the continuous annual and spatiotemporal changes were not analyzed. Future research needs to further improve the temporal precision of the analysis between consecutive years and seasonal water yield to obtain more refined research results.

The process of researching water production services is not without its inherent uncertainties. Firstly, the resolution of the raster has the potential to obscure minor differences in land use across a given area. Secondly, the InVEST model is inadequate in its consideration of the impacts of densely populated areas of human activities, policies such as water diversion and transfer, and upstream water inflow on actual water resources. Notwithstanding certain limitations and uncertainties, the InVEST model continues to be recognized by scholars at home and abroad as an exemplary instrument for evaluating ecosystem services [40].

5. Conclusions

• In 2000, 2010, and 2020, the water yield in Heilongjiang Province was 105.03 mm, 162.46 mm, and 269.34 mm, respectively. Compared to the year 2000, the water yield in 2020 increased by 164.31 mm (156.44%). From 2000 to 2020, the water yield showed an increasing trend, and the spatial distribution shows more in the east and less in the west. This distribution is highly consistent with the spatial distribution of precipitation. The water yield is highest in Jixi City and lowest in the Greater Khingan region area in Heilongjiang Province.
• From 2000 to 2020, the share of woodland was the highest, followed by farmland, while the share of built-up land was the lowest. The changes in farmland and unused land are significant. The increase in farmland was mainly due to the transfer of unused land and woodland, while the increase in unused land was mainly due to the transfer of waterbody. There has been a slight decrease in woodland and waterbody, with relatively small changes in grassland and built-up land areas.
• The order of water yield of land use/land cover types, from the greatest to the least, is as follows: built-up land, unused land, farmland, grassland, woodland, and waterbody. Therefore, policies such as urbanization and wetland protection will increase water yield, while policies to expand forest or grassland areas will decrease water yield.
• The climatic conditions are of paramount importance with regard to the yield of water. Specifically, from 2000 to 2020, climate change contributed as much as 99.58% to water yield, while changes in land use types contributed only 0.42%.
• Despite a decline in the overall population of Heilongjiang since 2010, the urban population continues to grow. This ultimately results in an indirect increase in water yield.