The impact of land use and land cover changes on East Asian summer ...

ATMOSPHERIC SCIENCE LETTERS Atmos. Sci. Let. (2017) Published online in Wiley Online Library (wileyonlinelibrary.com) DOI: 10.1002/asl.788

The impact of land use and land cover changes on East Asian summer monsoon precipitation using the WRF-mosaic approach Deming Zhao1* and Jian Wu2 1 CAS

Key Laboratory of Regional Climate-Environment for Temperate East Asia, Institute of Atmospheric Physics, Chinese Academy of Sciences, Beijing, China 2 Department of Atmospheric Science, Yunnan University, Kunming, China

*Correspondence to: D. Zhao, CAS Key Laboratory of Regional Climate-Environment for Temperate East Asia, Institute of Atmospheric Physics, Chinese Academy of Sciences, #40 Huayanli, Beijing 100029, China. E-mail: [email protected]

Received: 7 April 2017 Revised: 26 August 2017 Accepted: 28 September 2017

Abstract Based on satellite-derived land surface data, the impact of land use and land cover (LULC) changes between the 1980s and 2010s on the monsoon-related circulation and precipitation over East Asia was explored using the Weather Research and Forecasting (WRF) regional climate model with the mosaic approach, which could consider subgrid-scale LULC characteristics and the corresponding changes. Simulated results using the satellite-based LULC data show that the monsoon-related precipitation and rain belt movement could be well reproduced, whereas the precipitation was generally overestimated. In terms of the general effect of LULC changes between the 1980s and 2010s, the precipitation decreased in the north and increased in the south. Significant subregional characteristics are apparent in China, especially with respect to East Asian summer monsoon (EASM)-related precipitation. The increased roughness induced weakened near-surface wind speeds in the southern part of the EASM region, while the decreased roughness resulted in intensified values in the northern part. Meanwhile, the impacted EASM-related circulation and moisture flux, which could be well explained by the LULC changes induced, influenced the pressure gradient and temperature gradient at low to middle latitudes, resulting in a weakened moisture flux from the southwest and the southeast, whereas it intensified from the South China Sea and at middle to high latitudes. In general, with weakened moisture flux at low to high latitudes, the EASM-related rain belt’s northward movement was restrained. Although the enhanced westerly currents contributed to the increased moisture flux at middle to high latitudes, the precipitation there decreased due to the stronger weakened northward moisture flux. Keywords: land use and land cover changes; precipitation; East Asia summer monsoon; circulation; moisture flux

1. Introduction Although the impact of land use and land cover (LULC) changes on climate at global or regional scales might be low, the importance at local scales is worth considering (Findell et al., 2007), as LULC changes are sure to alter the local water exchange between the land surface and the atmosphere, thus having an effect on local, regional, and even global climate. Meanwhile, due to the new recognition of the interactions between the land surface and the atmosphere, and because there are more complex numerical models, with which the physical processes can be more objectively described, the understanding of the impacts of LULC changes has been improved. Earlier studies, for example, on the deforestation of the Amazon showed a negligible influence on the near-surface climate (Henderson-Sellers and Gornitz, 1984); however, with a finer-resolution model and more complex physical processes, the precipitation could increase by 30% (Henderson-Sellers et al., 1993). Since then, research in this field has developed

to include evaluations of the impacts of LULC changes at different spatial and temporal scales. When using a general circulation model at a global scale, it has been shown that precipitation could increase by 93% when the underlying surface changes from desert to forest (Fraedrich et al., 1999), whereas the global annual surface air temperature (SAT) could decrease by 0.13–0.25 ∘ C as a result of deforestation (Brovkin et al., 2006). Studies conducted by Sacks et al. (2009) showed that irrigation could significantly affect the regional climate. Indeed, many studies on the influence of LULC changes focused on regional scales, such as in the African Sahara (Charney, 1975), in northern forest regions (Bonan et al., 1992), and in tropical forest regions (Hasler et al., 2009). Meanwhile, the effects of the expansion of urban land cover on precipitation have also attracted considerable attention (Molders and Olson, 2004; Zhao and Wu, 2017a). From a climatic perspective, East Asia is a region that is primarily influenced by monsoonal processes (Fu et al., 2000). Therefore, the precipitation possesses

© 2017 The Authors. Atmospheric Science Letters published by John Wiley & Sons Ltd on behalf of the Royal Meteorological Society. This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.

D. Zhao and J. Wu

significant monsoon-related climate characteristics. Changes in the intensity of monsoon circulation and precipitation were affected by natural, and anthropogenic forcing including greenhouse effects, aerosol emissions (Menon et al., 2002), and LULC changes (Fu, 1997, 2003), among which the impact of LULC changes on a monsoon climate have been examined at different spatial and temporal scales. For example, the thermal difference between ocean and land is altered by changes in vegetation, which in turn affect the monsoon circulation (Chen et al., 2002). Vegetation degradation induces a strengthening of near-surface winds (Ding et al., 2005), and the altered circulation affects precipitation (Gao et al., 2007). The conversion of cropland to forest results in an increase in the soil water content, altering the exchange between the land surface and the atmosphere (Jiang et al., 2009). Fu and Yuan (2001) found that the destruction of natural vegetation is a possible contributing factor to the weakening of the East Asian summer monsoon (EASM), which is an important change in addition to natural variability (Fu, 2003). These studies were performed based on virtual tests or concentrated on the conversion of a certain land use category. Meanwhile, the impacts of LULC changes have not been explored using the fixed-in-time LULC data in the commonly used regional climate models, such as the Pennsylvania State University / National Center for Atmospheric Research mesoscale model (MM5) and WRF. Therefore, two new satellite images from the 1980s and 2010s, which could reveal LULC changes of the past 30 years over East Asia, are adopted here to explore the influence on EASM precipitation.

2. Experimental design and data 2.1. Experimental design Two numerical integrations based on the LULC data of the 1980s and 2010s (LU80 and LU10) using the WRF model were integrated for 36 years (1979–2014), in which the first 2 years were regarded as spin-up time, and 34 years (1981–2014) of results were analyzed. The model domain was as follows (as shown in Figure S1(a)): central latitude (35∘ N) and longitude (108.5∘ E), 289 and 229 grid points in the longitudinal and latitudinal directions, respectively (including a 10-point-grid buffer zone not used in the analysis), a 30-km horizontal resolution, 51 levels in the vertical direction, and 10 hPa at the top of the model. Meanwhile, the main physical parameterization schemes include (Skamarock et al., 2008): the WRF Single-Moment 6-class graupel microphysics scheme, the RRTMG shortwave and longwave radiation schemes with the greenhouse gas concentrations as the input (WRF modifications for regional climate simulation, CLWRF; Fita et al., n.d.), the unified Noah land-surface model with the mosaic approach (Li et al., 2013), which can consider the N (N equals 3) dominant land use categories instead of only one dominant land use category, the single-layer Urban Canopy

Model, the Yonsei University boundary-layer scheme, and the Kain-Fritsch (new Eta) cumulus scheme.

2.2. Data Detailed information regarding the acquisition of satellite-retrieved images for LU80 and LU10 to reveal the LULC changes over the past 30 years was presented in Yang et al. (2017) and Li et al. (2017) and is described in Text S1. The initial conditions and time-varying boundary conditions including sea surface temperature (updated every 6 h) for the integrations were taken from the NCEP/NCAR reanalysis dataset with a resolution of 2.5∘ × 2.5∘ , which have been adopted to perform studies on the impacts of urbanization using different regional climate models over Tokyo, Japan (Inamura et al., 2011), Chongqing, China (Zhang et al., 2008), and three city clusters over eastern China (Zhou et al., 2015), as well as regional climate simulations over East Asia (Zhao, 2013; Zhao and Wu, 2017a, 2017b). To evaluate the model’s performance with respect to precipitation simulations, the Global Precipitation Climatology Centre (GPCC) Monitoring Product, with a resolution of 0.5∘ × 0.5∘ (1981–2014, Schneider et al., 2015), was used as the observational precipitation dataset.

3. Results 3.1. Changes in roughness Changes in the roughness length for all and for three dominant land use categories in the summer (June–July–August, JJA), which were similar to the annual values (figures omitted) and could express LULC changes in the corresponding areas, are shown in Figure 1. Clear differences in the roughness between the first dominant (dominant approach) and all (mosaic approach, Li et al., 2013) land use categories, in which the contributions to all land use categories at a certain grid cell were 75% for the former and 98% for the latter, can be seen. These revealed that the contributions to total roughness from the subgrid-scale LULC and its changes could not be negligible and might have a further impact on the energy and water exchanges between the land surface and the atmosphere. Meanwhile, greater changes could be detected over eastern China (Table 1), in which the roughness generally increased in the south and decreased in the north. Changes in the roughness could be well interpreted by the LULC changes for the dominant three tiles. As the EASM precipitation was our concentration, LULC changes in Northeast China (NE), North China (NC), East China (EC), and South China (SC) (Figure S1(b)) over EASM regions were detailed analyzed. The LULC changes for the dominant three tiles (Figure S2) showed that quite different land use categories changes between NE and the other three subregions, in which changes mainly occurred from forests to croplands for the former (inducing decreased roughness), whereas

© 2017 The Authors. Atmospheric Science Letters published by John Wiley & Sons Ltd on behalf of the Royal Meteorological Society.

Atmos. Sci. Let. (2017)

The impact of LULC changes on EASM precipitation using mosaic approach

Figure 1. Changes in the JJA roughness length for (a) all and ((b): the first; c: the second; and c: the third) individual of the three dominant land use categories for LU10-LU80 (units: m). Table 1. Changes in the annual and JJA averaged values for the precipitation (units: mm day−1 ), roughness length (units: m), near-surface wind speed (units: m s−1 ), and latent heat flux (units: W m−2 ) over the EASM region from LU10-LU80. Annual

NE NC EC SC

JJA

Precipitation

Roughness

Near-surface wind speed (%)

Latent heat flux

Precipitation

Roughness

Near-surface wind speed (%)

Latent heat flux

−0.035 −0.019 −0.016 0.050

−0.032 0.016 0.042 0.047

2.0 −7.2 −4.9 −3.3

−0.60 −0.18 −0.66 −0.83

−0.068 −0.10 −0.098 0.31

−0.021 0.019 0.039 0.043

0.23 −6.4 −3.2 −3.1

−1.08 −1.30 −1.18 −1.44

from croplands and cropland/natural vegetation mosaic to forests (inducing increased roughness). Meanwhile, urban and built-up areas increased due to the decreased areas in forests, croplands and cropland/natural vegetation mosaic, especially for the second and third tiles in NC, EC, and SC, which resulted in increased roughness. Contributions to the total areal land use fractions for the dominant tile in the four subregions for LU80 and LU10 ran between 68.1 and 79.1%. The corresponding values ranged from 13.9 to 21.6% for the second dominant tile, while from 4.3 to 7.9% for the third dominant tile (Table S1). These meant that the second and third tiles (Table S2) might had a considerable influences on the energy and water exchanges between the land surface and atmosphere, thus had influences on EASM circulation and precipitation.

LinHo, 2002): the western North Pacific summer monsoon and the EASM (SEA, between 110∘ and 125∘ E), the Indian Summer Monsoon (SINDIA, to the west of 98∘ E), and the transitional zone covering the Indo-China Peninsula and the Yun-Gui Plateau (SINDO, between 98∘ and 110∘ E) were considered. The precipitation mainly occurred between May and September (rainy seasons), with lower precipitation occurring from October to April (dry seasons). The rain belt movement, which was highly affected by the monsoon circulation (Figure 2(a), Figures S3(a) and S3(d)), could generally be reproduced by the simulations, though with stronger intensity (Figure 2(b), Figures S3(b) and S3(e)).

3.2. Model performance with respect to precipitation simulation

Because of the varied precipitation intensity between the northern and southern parts of the simulated domain (weaker for the former and much stronger for the latter), relative values of the precipitation differences caused by LU10-LU80 (Text S2), which indicated that the changes that occurred during dry seasons were less substantial,

Latitude-month cross sections of the monthly mean precipitation, which correspond to the rain belt movements of three regional monsoon systems (Wang and

3.3. The impact on the precipitation



D. Zhao and J. Wu

Figure 2. Latitude-month cross sections of monthly mean precipitation from ((a), units: mm day−1 ) GPCC, ((b), units: mm day−1 ) LU80, and the corresponding differences from ((c), relative values, units: %) LU10-LU80 during SEA.

are discussed (Figure 2(c), Figures S3(c) and S3(f)). However, notable changes could be detected during rainy seasons. Over SEA, as the rain belt moved northward, the precipitation decreased in the north. However, as the rain belt retreated, the precipitation in southeastern China increased (Figure 2(c)). The changes in the annual precipitation for LU10-LU80 shown in Table 1 and Figure S4 further revealed marked subregional characteristics in China, with the largest values occurring over the EASM region, especially in EC and SC. Over SINDO, the precipitation north of 40∘ N became much weaker (Figure S3(c)). Distributions similar to those over SEA were found in terms of the influences on the northward movement and southward retreat of the rain belt. Whereas the changes in the precipitation Over SINDIA (Figure S3(f)) were quite different from those over SEA and SINDO.

3.4. Influence on EASM circulation Given that precipitation primarily occurs during JJA, the EASM-related circulation caused by LULC changes are discussed in detail. The zonal wind at 850 hPa over the EASM region revealed enhanced westerly winds in the north and weakened westerly and intensified easterly winds in the south (Figures 3(a) and (b)). Meanwhile, differences in the meridional winds were generally negative, showing weakened southerly winds, except for strengthened values from the South China Sea to southeastern China (Figures 3(c) and (d)). Consequently, for the EASM region, monsoon circulation was generally weakened in the south, which mainly resulted from decreased westerly and southerly winds (Figures 3(e) and (f)). However, in the north, the circulation was generally enhanced due to increased westerly and decreased southerly winds. Regarding the JJA temperature field at 850 hPa over the EASM region (Figure 4(a)), the temperature generally increased, with the exception of decreased values to the east of the Tibetan Plateau and the areas in northeast Asia. However, the decreased values in the north were much greater. The greater increased temperature at middle latitudes induced increased temperature gradients at middle to high latitudes and decreased temperature

gradients at low to middle latitudes, which resulted in enhanced westerly winds in the north and weakened westerly winds in the south. The influence of LULC changes on the JJA sea level pressure (SLP) over the EASM region, shown in Figure 4(b), revealed generally increased pressure at middle latitudes, which induced a decreased pressure gradient at low to middle latitudes and contributed to weakened southerly winds. Because Asia is controlled by low pressure during JJA, the much larger increases in pressure over the continent induced a decreased pressure gradient between the northwestern Pacific subtropical high and the continent, which resulted in weakened southeasterly winds. Meanwhile, the decreased SLP along the seashore of southeastern China contributed to the strengthened southerly winds. Regarding differences in precipitable water caused by LULC changes over the EASM region (Figure 4(c)), precipitable water increased in the south and decreased north of 35∘ N. This result was consistent with the weakened EASM circulation related to limited northward moisture transport. The impacts of LULC changes on the regional climate mainly focused on the land areas. However, this influence could have affection on the climate system through the upstream and downstream effect.

3.5. The impact on moisture flux The moisture transported in the EASM system originates from four processes: southwestern currents from the Bay of Bengal, southeastern currents from the northwestern Pacific subtropical high, southerly winds from the South China Sea, and westerly currents at middle to high latitudes. The moisture flux of the former two flows at 850 hPa was generally weakened, whereas that of the latter two was intensified (Figures 3(g) and (h)). Regarding the results throughout the troposphere, a distribution similar to that of the results at 850 hPa was found (Figures 3(i) and (j)).

4. Summary Using two satellite-derived images from the 1980s and 2010s, the impacts of LULC changes on EASM




(a)

(b)

(c)

(d)

(e)

(f)

(g)

(h)

(i)

(j)

Figure 3. JJA mean results for (a, c, e) LU80 and (b, d, f) the corresponding changes for LU10-LU80 winds (color scale) for (a, b) zonal (m s−1 ), (c, d) meridional (m s−1 ), and (e, f) total and vector (m s−1 ) at 850 hPa. JJA mean moisture flux (color scale) and vectors for (g, i) LU80 and (h, j) the corresponding changes for LU10-LU80 (g, h) at 850 hPa (10−3 kg hPa−1 m−1 s−1 ), and (i, j) across the entire troposphere (10−3 kg m−1 s−1 ).



D. Zhao and J. Wu

Figure 4. Changes in the JJA temperature field ((a), units: ∘ C) at 850 hPa, and the sea level pressure ((b), units: hPa) and precipitable water ((c), units: mm) from LU10-LU80.

precipitation were explored using the WRF model with the mosaic approach, which could consider subgrid-scale LULC changes. The roughness generally increased in the south and decreased in the north, which induced a decrease of near-surface wind speed over EASM region with the exception of an increase in NE. Due to the LULC changes, the precipitation decreased in the north of EASM region, but it increased in the south during the southward retreat of the rain belt. These characteristics could be partially explained by the influenced EASM circulation and moisture flux, which displayed different spatial pattern as the changes of near-surface wind speed due to the strong baroclinic nature of the middle latitude atmosphere: the weakened circulation in the south due to the decreased westerly and southerly winds, and the enhanced circulation in the north due to the increased westerly and decreased southerly winds, which could be further expressed by the influenced temperature fields at 850 hPa and SLP. The increased temperature gradient at middle to high latitudes and low to middle latitudes, respectively, contributed to the

enhanced westerly winds in the north and the weakened westerly winds in the south. Meanwhile, the decreased pressure gradient at low to middle latitudes resulted in weakened southerly winds, while the decreased gradient between the northwestern Pacific subtropical high and the continent resulted in weakened southeasterly winds. The weakened EASM circulation limited the northward moisture transport to northeastern China, which induced increased precipitable water in the south and decreased precipitable water in the north. The moisture flux from the south (including the southwestern and southeastern parts) was generally weakened, while it increased for southerly currents from the South China Sea and western currents at middle to high latitudes. Changes in the JJA latent heat flux over the EASM region were consistently negative, whereas the mechanism differed (Table 1). The JJA near-surface wind speeds over EASM region were generally weakened due to LULC changes, which were in agreement with the total effects from natural and anthropogenic forcing, as well as the results from Wu et al. (2016, 2017) over the east China Plain and Zha et al. (2016) across the whole China using the observation (observational data from China meteorological administration) minus reanalysis (ERA-Interim reanalysis data), which revealed that the increased drag coefficient from LUCC changes account for the long-term weakened near-surface wind speed. Changes in moisture flux over different subregions resulted from the influenced not only large-scale circulation related moisture flux, but also local upward moisture flux from the surface to the atmosphere. Due to LULC changes, the weakened EASM circulation prevented the northward water vapor transportation to the northern China and reserved more in SC, meanwhile, the local upward moisture fluxes over EASM areas were all decreased. In SC, though the local upward moisture flux decreased, the increased large-scale moisture flux was greater, which contributed to an increase of the precipitation. In EC and NC, both the local upward and large-scale moisture fluxes decreased, which contributed to the decreased precipitation there. In NE, the decreased local upward moisture flux, and the increased large-scale moisture from westerly flow, which was much weaker than that from EASM circulation-related water vapor transportation, contributed to the decreased precipitation there. As a result, the latent heat flux decreased in all four subregions over EASM areas and only the precipitation in SC increased. In NC and EC, with decreased near-surface wind speed and weakened precipitation, the latent heat flux decreased by 1.30 and 1.18 W m−2 , respectively. In NE, with decreased precipitation, the latent heat flux decreased by 1.08 W m−2 , though the near-surface wind speed slightly intensified. In SC, with decreased near-surface wind speed, the latent heat flux decreased by 1.44 W m−2 . The simulated precipitation over the EASM region has generally been overestimated, as revealed by previous studies (Zhao, 2013; Zhao and Wu, 2017b), which could be attributed to the enhanced simulated




EASM circulation and the corresponding moisture flux. Therefore, the decreased precipitation due to LULC changes could partially express the importance of LULC changes in long-term simulations over East Asia.

Acknowledgements This work was supported by the National Natural Science Foundation of China under grant no. 41775087, the National Key R&D Program of China under grant no. 2016YFA0600403, the National Natural Science Foundation of China under grant no. 41675149, the National Key Basic Research Program on Global Change under grant no. 2011CB952003, the Chinese Academy of Sciences Strategic Priority Program under grant no. XDA05090206, and the Jiangsu Collaborative Innovation Center for Climatic Change

Supporting information The following supporting information is available: Appendix S1 Supporting information on satellite-retrieved land use and land cover data and the impact on EASM precipitation. Text S1. Satellite-retrieved land use and land cover data from the 1980s and 2010s Text S2. The impact of LULC changes on the precipitation over SINDO and SINDIA Figure S1. (a) Model domain and topography distribution (unit: m), and (b) the eight subregions of China (NE: Northeast China; NC: North China; EC: East China; SC: South China; SW: Southwest China; NWE: eastern part of Northwest China; NWW: western part of Northwest China; TP: the Tibetan Plateau). Figure S2. Comparisons on different land use categories for the dominant three tiles (the first: a, d, g, j; the second: b, e, h, k; and the third: c, f, i, l) in (a-c) NE, (d-f) NC, (g-i) EC and (j-l) SC between LU80 and LU10. Figure S3. Latitude-month cross-sections of monthly mean precipitation from (a, d, units: mm day-1) GPCC, (b, e, units: mm day-1) LU80, and the corresponding differences from (c, f, relative values, units: %) LU10-LU80 during (a-c)SINDO and (d-f) SINDIA. Figure S4. Spatial distributions of JJA mean precipitation changes for LU10-LU80, with the cross lines denoting areas passing the t-test at the 90% confidence level (units: mm day-1 ). Table S1. Contributions to the total areal land use fractions for the dominant three tiles in the four subregions of EASM areas for LU80 and LU10 (units: %). Table S2. Grid numbers of land use categories for the domiant three tiles in the four subregions of EASM areas for LU80 and LU10.

References Bonan GB, Pollard D, Thompson SL. 1992. Effects of boreal forest vegetation on global climate. Nature 359: 716–718. Brovkin V, Claussen M, Driesschaert E, Fichefet T, Knicklighter D, Loutre MF, Matthews D, Ramankutty N, Schaeffer M, Sokolov A.

2006. Biogeophysical effects of historical land cover changes simulated by six Earth system models of intermediate complexity. Climate Dynamics 26: 587–600. Fu CB, Yuan HL. 2001. A virtual experiment on the effect of natural vegetation recovery to regional climate and environment over East Asia (in Chinese). Chinese Science Bulletin 46: 691–695. Charney J. 1975. Dynamics of deserts and drought in the Sahel. Quarterly Journal of the Royal Meteorological Society 101: 193–202. Chen X, Yu G, Liu J. 2002. Studies on the mechanism of Holocene climate simulation and its temperature variation in East Asia. Science China Earth Sciences 32: 335–345 (in Chinese). Ding YH, Li QP, Dong WJ. 2005. A numerical simulation study of the impacts of vegetation changes on regional climate in China. Acta Meteorologica Sinica 63: 613–621 (in Chinese). Findell KL, Stouffer RJ, Shevliakova E, Milly PCD. 2007. Modeled impact of anthropogenic land cover change on climate. Journal of Climate 20: 3621–3634. Fita L, Fernandez J, Garcia-Diez M. CLWRF: WRF modifications for regional climate simulation under future scenarios. http://meteo .unican.es/files/pdfs/WRFusers_clwrf.pdf. (accessed 13 October 2017) Fraedrich K, Kleidon A, Lunkeit F. 1999. A green planet versus a desert world: estimating the effect of vegetation extremes on the atmosphere. Journal of Climate 12: 3156–3163. Fu CB. 1997. Concept of "General monsoon system", an earth system science view on Asia monsoon. Proceedings of the International Workshop on Regional Climate Modeling of the General Monsoon System in Asia. Beijing, China, 1–6. Fu CB. 2003. Potential impacts of human-induced land cover change on East Asia monsoon. Global and Planetary Change 37: 219–229. Fu CB, Wei HL, Qian Y, Chen M. 2000. Documentation on regional integrated environmental model system (RIEMS version 1). TEACOM science reports START Regional Committee for temperate East Asia, TSR.2000. Gao XJ, Zhang DF, Chen ZX, Pal JS, Giorgi F. 2007. Numerical simulation for the impact of current land use on regional climate in China (in Chinese). Science China Earth Sciences 37: 397–404. Hasler N, Werth D, Avissar R. 2009. Effects of tropical deforestation on global hydroclimate: a multi-model ensemble analysis. Journal of Climate 22: 1124–1141. https://doi.org/10.1175/2008JCLI2157.1. Henderson-Sellers A, Gornitz V. 1984. Possible climatic impacts of land cover transformations, with particular emphasis on tropical deforestation. Climatic Change 6: 231–258. Henderson-Sellers A, Dickinson RE, Durbridge TB, Kennedy PJ, McGuffie K, Pitman AJ. 1993. Tropical deforestation: modeling local- to regional scale climate change. Journal of Geophysical Research 98: 7289–7315. Inamura T, Izumi T, Matsuyama H. 2011. Diagnostic study of the effects of a large city on heavy rainfall as revealed by an ensemble simulation: a case study of central Tokyo, Japan. Journal of Applied Meteorology and Climatology 50: 713–728. Jiang Y, Xu LP, Yang GH, Ren GX. 2009. Microclimate effects under different forest-grass rehabilitation of vegetation models in summer (in Chinese). Agricultural Research in the Arid Areas 25: 162–174. Li D, Bou-Zeid E, Barlage M, Chen F, Smith JA. 2013. Development and evaluation of a mosaic approach in the WRF-Noah framework. Journal of Geophysical Research: Atmosphere 118: 11918–11935. Li HX, Xiao PF, Feng XZ, Yang YK, Wang LX, Zhang WB, Wang XH, Feng WD, Chang X. 2017. Using land long term data records to map land cover changes in China over 1981–2010. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing 10: 1372–1389. https://doi.org/10.1109/JSTARS.2016.2645203. Menon S, Hansen JE, Nazarenko L, Luo YF. 2002. Climate effects of black carbon aerosols in China and India. Science 297: 2250–2253. Molders N, Olson MA. 2004. Impact of urban effects on precipitation in high latitudes. Journal of Hydrometeorology 5: 409–429. Sacks WJ, Cook BI, Buenning N, Levis S, Helkowski JH. 2009. Effects of global irrigation on the near-surface climate. Climate Dynamics 33: 159–175. Schneider U, Becker A, Finger P, Meyer-Christoffer A, Rudolf B, Ziese M. 2015. GPCC Full Data Reanalysis Version 7.0 at 0.5∘ : Monthly Land-Surface Precipitation from Rain-Gauges built on GTS-based



D. Zhao and J. Wu

and Historic Data. https://doi.org/10.5676/DWD_GPCC/FD_M_ V7_050. ftp://ftp.dwd.de/pub/data/gpcc/html/fulldata_v7_doi_down load.html (accessed 13 October 2017) Skamarock WC, Klemp JB, Dudhia J, Gill DO, Barker DM, Duda MG, Huang X-Y, Wang W. Powers JG. 2008. A description of the advanced research WRF Version 3. NCAR Technical Note, NCAR/TN–475+STR. Wang B, LinHo. 2002. Rainy season of the Asian–Pacific summer monsoon. Journal of Climate 15: 386–398. Wu J, Zha JL, Zhao DM. 2016. Estimating the impact of the changes in land use and cover on the surface wind speed over the East China Plain during the period 1980-2011. Climatic. Dynamics (Pembroke, Ont.) 46: 847–863. https://doi.org/10.1007/s00382-015-2616-z. Wu J, Zha JL, Zhao DM. 2017. Evaluating the effects of land use and cover change on the decrease of surface wind speed over China in recent 30 years using a statistical downscaling method. Climate Dynamics 48: 131–149. https://doi.org/10.1007/s00382-016-3065-z. Yang YK, Xiao PF, Feng XZ, Li HX. 2017. Accuracy assessment of seven global land cover datasets over China. ISPRS Journal of Photogrammetry and Remote Sensing 125: 156–173.

Zha JL, Wu J, Zhao DM. 2016. Effects of land use and cover change on the near-surface wind speed over China in the last 30 years. Progress in Physical Geography 41: 46–67. https://doi.org/10.1177/ 0309133316663097. Zhang HB, Hanaki K, Sato N, Izumi T, Aramaki T. 2008. Modified RAMS-urban canopy model for heat island simulation in Chongqing, China. Journal of Applied Meteorology and Climatology 47: 509–524. Zhao DM. 2013. Performance of Regional Integrated Environment Modeling System (RIEMS) in precipitation simulations over East Asia. Climate Dynamics 40: 1767–1787. Zhao DM, Wu J. 2017a. The influence of urban surface expansion in China on regional climate. Journal of Climate 30: 1061–1080. Zhao DM, Wu J. 2017b. Inclusion of land use changes in long-term regional climate simulations over East Asia. Atmospheric Science Letters 18: 187–192. https://doi.org/10.1002/asl.744. Zhou L, Jiang ZH, Li ZX, Yang XQ. 2015. Numerical simulation of urbanization climate effects in regions of East China (in Chinese). Chinese Journal of Atmospheric Sciences 39: 596–610.

