Modelling the effects of climate change on transpiration and evaporation in natural and constructed grasslands in the semi-arid Loess Plateau, China

https://doi.org/10.1016/j.agee.2020.107077Get rights and content

Highlights

  • Hydrus-1D was used to simulate T and E of three grasslands in the Loess Plateau.

  • The effects of climate change on T were greater than that on E.

  • Precipitation was the most dominant climatic factor affecting T and E.

  • Constructed grasslands had higher water use efficiency than natural one.

  • Constructed grasslands led to worse soil water conditions than natural one.

Abstract

Investigating the effects of climate change on transpiration (T) and evaporation (E) in natural and constructed grasslands is of theoretical and practical significance for vegetation restoration and water resource management in the Loess Plateau, China. In this study, a two-year field experiment was conducted in one natural (Imperata cylindrica plot) and two constructed grasslands (Pennisetum giganteum and Medicago sativa plots) in the semi-arid Loess Plateau. Hydrus-1D models were then established and validated to simulate T and E processes under scenarios combining climate characteristics in both future periods and different hydrological years. The results showed that under all climate scenarios, T in the natural grassland, and P. giganteum and M. sativa plots ranged from 63.7–179.3, 126.8–245.5, and 96.9–243.1 mm, respectively, while E ranged from 90.6–131.9, 68.7–92.8, and 70.4–92.4 mm, respectively. Both T and E showed a decreasing trend in the future periods, exhibiting the highest values in wet years and the lowest values in dry years. The effects of climate change on T were greater than that on E in all three grasslands. Of the studied grasslands, T of M. sativa plot exhibited the strongest response, followed by that in the natural grassland and P. giganteum plot. However, E of the natural grassland responded to the strongest degree, followed by that of the P. giganteum and M. sativa plots. Precipitation during the growing-season was the most dominant climatic factor affecting T and E, and was positively linearly related with T and E (P < 0.01) in all three grasslands. The conversion thresholds of precipitation for T-E dominance were 435.8, 149.4, and 253.3 mm for the natural grassland, and P. giganteum and M. sativa plots, respectively. In constructed grasslands, more water was lost through T than through E; hence, the T/ET ratios were high (>0.55). In contrast, T was high in the natural grassland only in conditions of sufficient precipitation; hence, the T/ET ratio was relatively lower (around 0.5). Precipitation distribution strongly affected the soil water supply, and the natural grassland was more efficient in maintaining soil water than the constructed grasslands. Our study not only improves our understanding of the hydrological processes and water budget under different grass restoration measures, but also provides valuable guidelines for the management of water resources and the restoration of ecological environment in the Loess Plateau.

Introduction

Evapotranspiration (ET) is an essential component of the soil-plant-atmosphere continuum (SPAC) system (Sprenger et al., 2017; Tello-García et al., 2020), and plays a crucial role as the driving force for hydrological cycling and energy balance in terrestrial ecosystems (Fisher et al., 2017). ET consists of two parts, namely vegetation transpiration (T) and soil evaporation (E). T is an important part of soil water consumption, and is regarded as productive water loss, because it is involved in essential plant physiological processes, and is thus closely associated with biomass production. In contrast, E is generally considered as non-productive water loss, since it is lost directly from the soil (Aouade et al., 2016; Unkovich et al., 2018). Therefore, quantifying T and E, and subsequently promoting the transformation from E to T without increasing the total amount of ET, has become an important approach to improve water use efficiency, and has generated worldwide research interests, especially in arid and semi-arid ecosystems (Aouade et al., 2016; Graham et al., 2016).

Both T and E are controlled by the combined effects of multiple factors, such as climatic factors, topography characteristics, vegetation species, and anthropogenic management (Odongo et al., 2019; Sadeghi et al., 2017; Zhou et al., 2006). Of these, climate change is generally believed to significantly influence the total ET in terrestrial ecosystems, which has been widely investigated at large spatial scales, including global, regional, and catchment scales (Allen et al., 2015; Odongo et al., 2019; Tello-García et al., 2020). In addition, several studies have been conducted taking specific vegetation properties into account, to improve our understanding of the effects of climate change on ET dynamics under diverse vegetation conditions (Alemayehu et al., 2017; Siad et al., 2019), and have significantly enriched our knowledge in the field. However, ET partitioning is still not entirely understood, owing to the high heterogeneity of surface conditions (topographic characteristics, vegetation species, soil types, etc.) (Montaldo et al., 2020). In recent years, studies have also been conducted on the effects of climate change on ET at smaller scales, aiming to provide experimental data and theoretical guidance for accurate large-scale simulations, and to promote more effective vegetation management (Allen et al., 2015; Chebbi et al., 2018). Numerous studies, focused particularly on T processes, have consistently concluded that climate change-induced evaporative demand promotes T, and that their relationships can be expressed as linear or non-linear functions in different climatic areas (Ghimire et al., 2014; Grossiord et al., 2017). In contrast, limited research has been conducted on the effects of climate change on E (Chebbi et al., 2018; Montaldo et al., 2020), and consequently, different effects of climate change on T and E, remain poorly understood to date. In addition, the different changes of T and E will lead to different eco-hydrological effects (e.g., alterations in water use efficiency and soil water conditions). Therefore, the lack of systematic studies on E will also result in the inadequate understanding of them. Moreover, considering the manpower and time required for long-term field experiments, little research has been designed to systematically investigate the effects of climate in different hydrological years on T and E (Ferlan et al., 2016; Montaldo et al., 2020). However, such information provides the basis for developing policies for vegetation conservation and water resource management, which further necessitates in-depth research on T and E under changing climatic conditions.

The Loess Plateau (6.4 × 105 km2) situated in northwest China is well-known for its severe soil erosion, which is considered to be caused by the combined effects of loose soil, intensive rainfall, and low vegetation coverage (Feng et al., 2018; Li et al., 2016; Zheng, 2006). During the past decades, the soil erosion has caused serious land degradation, and therefore, greatly hindered the agricultural production and ecosystem restoration (Feng et al., 2018; Zhou et al., 2006). In order to mitigate the effects of soil erosion and land degradation, measures of the ‘Engineering construction’ (Li et al., 2016) and the ‘Comprehensive erosion control’ (Shi and Shao, 2000) have been implemented in 1950s and 1960s, respectively. Furthermore, the Chinese central government launched the 'Grain to Green' project in 1999 to improve the vegetation cover and remediate ecological functions (Liu et al., 2019; Liu and Liu, 2018). Constructed vegetation, including trees and grass, significantly altered hydrological processes through ET, and greatly alleviated soil loss, as expected (Liu and Liu, 2018). However, it also had some negative impacts on the local ecosystems and the environment, such as soil desiccation due to excessive water uptake from deep soil layers by the roots and low survival rate of vegetation due to water deficit (Feng et al., 2016). These effects, which appeared to be the result of an imbalance between vegetation water demand and soil water supply, have already seriously hindered sustainable development of local ecosystems (Liu et al., 2019; Feng et al., 2016). Furthermore, the climatic trend of lower precipitation and higher temperature in the Loess Plateau also exacerbates this imbalance, which is highly detrimental to vegetation growth (Chebbi et al., 2018; Unkovich et al., 2018), especially in conditions of extreme drought (Allen et al., 2015; Raz-Yaseef et al., 2010; Tello-García et al., 2020). Hence, there is an urgent need to identify and compare the responses of T and E to the changing climate, under natural and constructed vegetation conditions, which would be pivotal for the selection of constructed vegetation species, improvement of their survival rates, and ecosystem restoration (Chebbi et al., 2018; Unkovich et al., 2018).

Nowadays, several approaches including water or energy balance theories (Kool et al., 2014), sap flow technologies (Chebbi et al., 2018), eddy-covariance measurements (Aouade et al., 2016), remote sensing technologies (Liu et al., 2017), and modellings (Chattaraj et al., 2014; Nosetto et al., 2012) are available to investigate the response of T or E to changes in climate. Of them, modellings showed strong applicability and flexibility in producing T and E dynamics at diverse temporal and spatial scales for various vegetation types, and were thus widely used (Aggarwal et al., 2017; Chattaraj et al., 2014). These relevant studies have, as yet, mainly been conducted in farmlands (Chattaraj et al., 2014), whereas have seldom been performed in grasslands. Besides, research comparing natural and constructed grasslands is relatively lacking. However, because of the relatively lower water consumption than tree species and advantages in being used as feed for livestock, grasslands have been highly recommended and are thus widely distributed in the Loess Plateau. Therefore, the eco-hydrological effects caused by different responses of their T and E to diverse climatic conditions should also be paid full attention to.

The present study was undertaken taking these points into consideration. A two-year field experiment was conducted in a natural (Imperata cylindrica plot, natural restoration without any anthropogenic management) and two typical constructed grasslands (Pennisetum giganteum and Medicago sativa plots, which were harvested and stored as feed for livestock after the growing season), in the semi-arid Loess Plateau, to establish models for simulating their T and E processes under different climatic scenarios (a combination of climate characteristics in both future periods and different hydrological years). The main objectives of this study were to: (1) identify the different effects of climate change on T and E based on ET partitioning, (2) investigate the eco-hydrological effects (water consumption strategies and soil water conditions) resulting from the different responses of T and E to the changing climate, and (3) compare the above characteristics in natural and constructed grasslands. The results of this study will improve our understanding of the hydrological processes occurring in the grasslands and their responses to climate change, and provide theoretical and practical guidance for grass species selection, grassland ecosystem conservation, and water resource management in the Loess Plateau.

Section snippets

Site description and experimental plot selection

Field experiments were conducted in the Nanxiaohegou Basin (107°30′–107°37′E and 35°41′–35°44′N, 1058–1450 m a.s.l.), Qingyang city, Gansu province in central Loess Plateau of China (Fig. 1). The basin was selected as a typical small-scale basin (36.5 km2) by the Yellow River Conservancy Committee in 1951 because of its typicality and representativeness in terrain characteristics and vegetation types. This region has a warm temperate continental climate, with characteristic hot and rainy

Results of parameter sensitivity analysis and calibration

The modified Morris screen method was used to calculate the specific values of sensitivity indices for soil hydraulic and vegetation physiological parameters (Table 1), and the results showed that the pore size distribution parameter (0–40 cm) was the most sensitive among all parameters, presenting a high sensitivity. Soil saturated moisture contents (0–40 and 40–100 cm), parameter of pore size distribution (40–100 cm), field water capacity, and all vegetation parameters were sensitive,

Reasons for the different responses of the grass species to climate change

Previous studies conducted worldwide have shown that natural and constructed grasslands have different ET characteristics (Gu et al., 2018; Huang et al., 2019; Li et al., 2019). An important reason was that T and E responded differently to their environmental controlling elements (Li et al., 2019), namely evaporative demand and water supply (Jiao et al., 2018; Montaldo et al., 2020). In reality, both T and E in many semi-arid areas have experienced a suppression due to the increasingly warm and

Conclusion

In this study, the effects of climate change on transpiration and evaporation of a natural (Imperata cylindrica plot) and two constructed grasslands (Pennisetum giganteum and Medicago sativa plots) in the Loess Plateau were investigated using the Hydrus-1D model based on pre-set climate scenarios. The results revealed that transpiration responses in all three grasslands were greater than that of evaporation under changing climate, and their transpiration response degrees could be ranked from

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

We sincerely thank the academic editor and anonymous reviewers for their insightful and constructive comments. We also thank the staff at Xifeng Experiment Station of Soil and Water Conservation as well as Xianghua Feng, Ling Zhang, Zhixu Zhang, Chong Fu, and Shaona Wang from Xi’an University of Technology, China, and instructor Sun, Mingming Guo, Hongliang Kang, and Qianhua Shi from Northwest A&F University, China, and Guangchen Chu from the Institute of Earth Environment, Chinese Academy of

References (81)

  • Chunjie Gu et al.

    Partitioning evapotranspiration using an optimized satellite-based ET model across biomes

    Agric. For. Meteorol.

    (2018)
  • D. Han et al.

    Hydroclimatic response of evapotranspiration partitioning to prolonged droughts in semiarid grassland

    J. Hydrol.

    (2018)
  • Z. Huang et al.

    Natural grasslands maintain soil water sustainability better than planted grasslands in arid areas

    Agric. Ecosyst. Environ.

    (2019)
  • X. Jia et al.

    Spatial variations in soil-water carrying capacity of three typical revegetation species on the Loess Plateau

    China. Agric. Ecosyst. Environ.

    (2019)
  • L. Jiao et al.

    Evapotranspiration partitioning and its implications for plant water use strategy: evidence from a black locust plantation in the semi-arid Loess Plateau, China

    For. Ecol. Manage.

    (2018)
  • D. Kool et al.

    A review of approaches for evapotranspiration partitioning

    Agric. For. Meteorol.

    (2014)
  • X. Li et al.

    A simple and objective method to partition evapotranspiration into transpiration and evaporation at eddy-covariance sites

    Agric. For. Meteorol.

    (2019)
  • J. Li et al.

    Modeling the effects of parameter optimization on three bioretention tanks using the HYDRUS-1D model

    J. Environ. Manage.

    (2018)
  • Z. Liu et al.

    Biophysical effect of conversion from croplands to grasslands in water-limited temperate regions of China

    Sci. Total Environ.

    (2019)
  • Y. Luo et al.

    Seasonal groundwater contribution to crop-water use assessed with lysimeter observations and model simulations

    J. Hydrol.

    (2010)
  • N. Montaldo et al.

    Fixed and variable components of evapotranspiration in a Mediterranean wild-olive - grass landscape mosaic

    Agric. For. Meteorol.

    (2020)
  • J.E. Nash et al.

    River flow forecasting through conceptual models part I - a discussion of principles

    J. Hydrol.

    (1970)
  • M.D. Nosetto et al.

    The hydrologic consequences of land cover change in central Argentina

    Agric. Ecosyst. Environ.

    (2012)
  • V.O. Odongo et al.

    Impact of land use and land cover transitions and climate on evapotranspiration in the Lake Naivasha Basin, Kenya

    Sci. Total Environ.

    (2019)
  • K.C. Perera et al.

    Comparison of hourly and daily reference crop evapotranspiration equations across seasons and climate zones in Australia

    Agric. Water Manag.

    (2015)
  • G.Y. Qiu et al.

    Experimental studies on the effects of the “Conversion of Cropland to Grassland Program” on the water budget and evapotranspiration in a semi-arid steppe in Inner Mongolia

    China. J. Hydrol.

    (2011)
  • G.Y. Qiu et al.

    Characteristics of soil evaporation, plant transpiration and water budget of Nitraria dune in the arid Northwest China

    Agric. For. Meteorol.

    (2015)
  • N. Raz-Yaseef et al.

    Effects of spatial variations in soil evaporation caused by tree shading on water flux partitioning in a semi-arid pine forest

    Agric. For. Meteorol.

    (2010)
  • N. Raz-Yaseef et al.

    Vulnerability of crops and native grasses to summer drying in the U.S. Southern Great Plains

    Agric. Ecosyst. Environ.

    (2015)
  • M.G. Schaap et al.

    Rosetta v1.2: A computer program for estimating soil hydraulic parameters with hierarchical pedotransfer functions

    J. Hydrol.

    (2001)
  • H. Shi et al.

    Soil and water loss from the Loess Plateau in China

    J. Arid Environ.

    (2000)
  • S.M. Siad et al.

    A review of coupled hydrologic and crop growth models

    Agric. Water Manag.

    (2019)
  • E. Tello-García et al.

    Drought- and heat-induced shifts in vegetation composition impact biomass production and water use of alpine grasslands

    Environ. Exp. Bot.

    (2020)
  • A. Trepekli et al.

    Seasonal evapotranspiration, energy fluxes and turbulence variance characteristics of a Mediterranean coastal grassland

    Agric. For. Meteorol.

    (2016)
  • M. Unkovich et al.

    Field measurements of bare soil evaporation and crop transpiration, and transpiration efficiency, for rainfed grain crops in Australia – a review

    Agric. Water Manag.

    (2018)
  • H. Wang et al.

    Modelling the effects of land cover and climate change on soil water partitioning in a boreal headwater catchment

    J. Hydrol.

    (2018)
  • B. Xi et al.

    Characteristics of fine root system and water uptake in a triploid Populus tomentosa plantation in the North China Plain: implications for irrigation water management

    Agric. Water Manag.

    (2013)
  • X. Yi et al.

    Comparison of three methods to develop pedotransfer functions for the saturated water content and field water capacity in permafrost region

    Cold Reg. Sci. Technol.

    (2013)
  • T. Zha et al.

    Interannual variation of evapotranspiration from forest and grassland ecosystems in western canada in relation to drought

    Agric. For. Meteorol.

    (2010)
  • F.L. Zheng

    Effect of vegetation changes on soil Erosion on the Loess Plateau

    Pedosphere

    (2006)
  • Cited by (30)

    View all citing articles on Scopus
    View full text