Changes in Local Growing Season Indices in Shanghai Due to Urbanization during 1873–2013

XIA Jiang-Jiang, YAN Zhong-Wei, and XIONG Zhe

Key Laboratory of Regional Climate-Environment for East Asia, Institute of Atmospheric Physics, Chinese Academy of Sciences, Beijing 100029, China

Changes in Local Growing Season Indices in Shanghai Due to Urbanization during 1873–2013

XIA Jiang-Jiang, YAN Zhong-Wei, and XIONG Zhe*

Key Laboratory of Regional Climate-Environment for East Asia, Institute of Atmospheric Physics, Chinese Academy of Sciences, Beijing 100029, China

The contribution of urbanization to the changes in growing season indices (e.g., the start date of the growing season, Ds) in Shanghai, eastern China, for the period 1873-2013 is assessed. The urban-related Ds advancement (referred to as the urbanization contribution, UC) is reconstructed based on the changing urban-land fraction within a grid box around Shanghai for 1873-2013. After removing the UC from the raw Ds time series, the secular trend (ST) representing the long-term climatic warming and multidecadal variability (MDV) representing an oscillatory component with a period of several decades are obtained by using the Ensemble Empirical Mode Decomposition (EEMD) method. Results show that the UC is comparable with the contribution of the ST for the period 1873-2013, when Ds exhibits an advancing rate of about 0.44 days per decade. MDV plays an important role in regulating changes in Ds on the decadal timescale, e.g., with a contribution of about 56.1% for 1981-2013.

growing season, urbanization, multidecadal variability, climate changes

1　Introduction

As important climatic indicators, growing season indices are useful for climatological and ecological applications (White et al., 1999; Robeson, 2002; Linderholm, 2006). In China, the thermal growing season, which is defined based on long-term meteorological observations (e.g., temperature records), has been extending for the last several decades, mainly due to an earlier onset of the growing season (Xu and Ren, 2004; Liu et al., 2010; Song et al., 2010; Yang et al., 2013). Christidis et al. (2007) suggested that, during that time, the lengthening of the growing season in Northern Hemisphere lands can be attributed to the anthropogenic forcings that have acted on the climate system. Xia et al. (2013) argued that the multidecadal variability (MDV) in the time series of growing season indices is associated with the Atlantic Multidecadal Oscillation (AMO). It accelerated the lengthening rate by about 53% during the period 1980-2009.

It is worth noting that, due to the rapid developmentduring the last few decades, urbanization has induced biases in observed local temperature time series (Ren et al., 2007; Wang et al., 2012), especially in eastern China (Zhao et al., 2014). Yang et al. (2013) indicated that one tenth of the increase in growing season degree days in eastern China for the period 1961-2007 can be attributed to urbanization effects. However, the urbanization-induced changes in the growing season for the whole 20th century has not been well studied.

This paper aims to assess the contribution of urbanization to the changes in growing season indices (e.g., the start date of the growing season). In addition, in order to obtain a more comprehensive result, the relative contributions of MDV and long-term anthropogenic warming are also analyzed. Section 2 describes the data and methods used in this study. Section 3 presents the results obtained. Section 4 provides further discussion and summarizes the key conclusions.

2　Data and method

The Shanghai daily surface air temperature series for the period 1873-2013 has been updated from the homogenized series developed by Yan et al. (2001) and Li and Yan (2009). Xia et al. (2015) suggested that the temperature of Shanghai is regionally (eastern China) representative. The local growing season (LGS) is defined using the local climatological annual mean temperature (Tm) as the temperature threshold (Christidis et al., 2007; Xia and Yan, 2014). Following this, Tmis defined as the average temperature over the period 1873-2013. For each year, the start date (Ds) of the LGS is the date when local temperature rises above Tmin the spring (in April).

The long-term AMO index series for 1901-2013 was obtained from http://www.esrl.noaa.gov/psd/data/timeseries/ AMO/.

The changing urban-land fraction within a grid box around Shanghai ((30-33°N, 119-122°E); Fig. 1a) for 1873-2013, as used in the Coupled Model Intercomparison Project Phase 5 (CMIP5) historical climate experiments, was obtained from http://tntcat.iiasa.ac.at:8787/ RcpDb/dsd?Action=htmlpage&page=welcome#.

The Ensemble Empirical Mode Decomposition (EEMD) method is used to decompose a complicated data series into a small number of amplitude-frequency-modulated oscillatory components of different timescales (Huang and Wu, 2008; Wu and Huang, 2009). The procedure of Wu et al. (2011) is adopted here to decompose the Dsseries into three parts: a secular nonlinear trend (ST) representing the long-term climatic warming; MDV representing those oscillatory components with timescales of decades up to a century; and high-frequency variability (HFV) as a sum of the shorter-term components.

3　Results

Xia and Yan (2014) found that Ds has a good linear relationship with April temperature. They suggested that for a 1°C warming of April temperature there should be an advancement of Ds by 5.2 days for Shanghai (Xia and Yan, 2014; Fig. 1a). Furthermore, there have been a wide range of estimates of the urbanization effect in China. Averaged over large areas, i.e., China, the urban contribution is about 10%, according to Jones et al. (2008). However, Yang et al. (2011) suggested that over metropolis-based stations (e.g., Shanghai) the urban contribution could be 44%. In fact, different methods may result in different estimates of the contribution of urbanization (Du et al., 2007; Jones et al., 2008; Yang et al., 2011; Wang et al., 2013). To facilitate discussion, we adopt the suggestion from a recent study by Zhao et al. (2014) that urbanization could have accounted for up to 9%-24% of the local warming trend in eastern China during 1951-2010.

In this study, we take the changing urban-land fraction (Fig. 1a) as the proxy of urbanization progress. We assume a linear relationship between the urban-land fraction and its contribution to advancing Ds, which allows us to roughly reconstruct the urban-related advancement in Ds for Shanghai, as shown in Fig. 1b. As shown in this figure, the reconstructed urban effect for Ds in Shanghai was about nil in 1873 and about -1.7 days (UC09) to -4.6 days (UC24) in 2013. The urban signal is somehow overwhelmed in strong interannual fluctuations of Ds and only visible in Fig. 1c after the 1980s.

The three timescale components of the Ds/Ds_ UC09/Ds_UC24 series are obvious, via EEMD (Fig. 2). Removing the urban effect (UC09 or UC24) makes the ST slighter (Fig. 2c), but barely influences the HFV (Fig. 2a) and MDV (Fig. 2b) inherent in the time series. Note that the AMO is roughly out of phase with the MDV of Ds variations in eastern China (Fig. 2b). It has been pointed out in previous work that the AMO plays a key role in regulating temperature changes (including April temperature) within a time window of decades for regions like East Asia (Lu et al., 2006; Li et al., 2009; Xia et al., 2013; Gao et al., 2015). Physically, our previous study (Xia et al., 2013) showed that the MDV component of the Ds in many regions over the world (including Shanghai, eastern China) has strong links with the AMO. This was explained by analyzing differences in atmospheric circulation between the warm and cool phases of the AMO. Wu et al. (2011) suggested that the ST, the last EEMD component, could reflect the anthropogenic warming signal. Thus, in this study, there is confidence in the decomposition results using the EEMD method for the low frequency signal, for both MDV and ST.

Table 1 shows the linear trends of different timescale components of Ds over different temporal spans. For the period 1873-2013, linear trends of UC (UC09 and UC24) range from -0.10 to -0.26 days per decade, indicating a 22.7% to 59.1% UC to the raw Ds time series (-0.44 days per decade). For the periods 1901-2013, 1951-2013, and 1981-2013, the UC to the total advancement of Ds is 16.9%-46.5%, 9.4%-25%, and 9.7%-25.9%, respectively. Based on the results in Table 1, we can also calculate the contributions of ST to the raw Ds time series, which are about 11.4%-54.6%, 14.1%-50.7%, 8.9%-25%, and 6.7%-17.7% for the periods 1873-2013, 1901-2013, 1951-2013, and 1981-2013, respectively. It is clear that the UC and ST is comparable for different periods and that ST and UC together play a key role in advancing theDs of the growing season on the century time scale. In comparison, MDV has a slightly negative contribution on the century time scale, i.e., for 1873-2013 and 1901-2013. However, it plays an important role in regulating changes in Ds on the decadal time scale, e.g., with a contribution of about 56.1% for 1981-2013.

Table 1 Linear trends of the raw time series, the urbanization contribution (UC), the high-frequency variability (HFV), the multidecadal variability (MDV), and the secular trend (ST) of the start date of the growing season (Ds) over different temporal spans. Values in square brackets indicate the linear trends with the UC being 9% and 24% for the period 1951-2010. Units: days per decade.

4　Discussion and conclusion

The present results highlight the specific need of regional attribution studies of changing climate extremes. In particular, for a region such as eastern China, regional development decision-makers are interested in how much a regional process such as urbanization, which local decision-makers can control, could enhance the risk of climate extremes in the region. However, the pattern of the effect of urbanization is compounded with that of greenhouse gas-associated global anthropogenic forcing. Therefore, the effect of local urbanization should be carefully considered in attribution studies of large-scale climate change. This paper has revealed the relative contributions of urbanization, MDV, and anthropogenic warming to the changes in Ds as recorded at Shanghai station. For the period 1873-2013, Ds exhibited an advancing rate of about 0.44 days per decade, mainly due to ST and UC. MDV plays an important role in regulating changes in Ds on the decadal scale.

Acknowledgements. This work was supported by the National Natural Science Foundation of China (Grant Nos. 41475078, 41405082, 51339004, and 91425304).

References

Christidis, N., P.-A. Stott, S. Brown, et al., 2007: Human contribution to the lengthening of the growing season during 1950-99, J. Climate, 20, 5441-5454.

Du, Y., Z.-Q. Xie, Y. Zeng, et al., 2007: Impact of urban expansion on regional temperature change in Yangtze River Delta, J. Geophys. Res., 17(4), 387-398.

Gao, L.-H., Z.-W. Yan, and X.-W. Quan, 2015: Observed and SST-forced multidecadal variability in Global Land surface air temperature, Climate Dyn., 44, 359-369.

Huang, N.-E., and Z. Wu, 2008: A review on Hilbert-Huang transform: Method and its applications to geophysical studies, Rev. Geophys., 46, RG2006, doi:10.1029/2007RG000228.

Jones, P.-D., D.-H. Lister, and Q. Li, 2008: Urbanization effects in large-scale temperatures with an emphasis on China, J. Geophys. Res., 113, D16122, doi:10.1029/2008JD009916.

Li, S.-L., Y.-M. Wang, and Y.-Q. Gao, 2009: A review of the researches on the Atlantic Multidecadal Oscillation (AMO) and its climate influence, Trans. Atmos. Sci. (in Chinese), 32(3), 458-465.

Li, Z., and Z.-W. Yan, 2009: Homogenized daily mean/maximum/ minimum temperature series for China from 1960-2008, Atmos. Oceanic Sci. Lett., 2(4), 1-7.

Linderholm, H.-W., 2006: Growing season changes in the last century, Agric. Forest Meteor., 137, 1-14.

Liu, B.-H., M. Henderson, Y.-D. Zhang, et al., 2010: Spatiotemporal change in China's climatic growing season: 1955-2000, Climatic Change, 99, 93-118.

Lu, R., B. Dong, and H. Ding, 2006: Impact of the Atlantic Multidecadal Oscillation on the Asian summer monsoon, Geophys. Res. Lett., 33, L24701, doi:10.1029/2006GL027655.

Ren, G., Z. Chu, Z. Chen, et al., 2007: Implications of temporal change in urban heat island intensity observed at Beijing and Wuhan stations, Geophys. Res. Lett., 34, L05711, doi:10.1029/2006GL027927.

Robeson, S.-M., 2002: Increasing growing-season length in Illinois during the 20th century, Climatic Change, 52, 219-238.

Song, Y.-L., H.-W. Linderholm, D. Chen, et al., 2010: Trends of the thermal growing season in China, 1951-2007, Int. J. Climatol., 30, 33-43.

Wang, J., J. Feng, Z. Yan, et al., 2012: Nested high-resolution modeling of the impact of urbanization on regional climate in three vast urban agglomerations in China, J. Geophys. Res., 117, D21103, doi:10.1029/2012JD018226.

Wang, J., Z.-W. Yan, P. D. Jones, et al., 2013: On "observed minus reanalysis" method: A view from multidecadal variability, J. Geophys. Res., 118, 1-9.

White, M.-A., S.-W. Running, and P.-E. Thornton, 1999: The impact of growing-season length variability on carbon assimilation and evapotranspiration over 88 years in the eastern US deciduous forest, Int. J. Biometeorol., 42, 139-145.

Wu, Z., and N.-E. Huang, 2009: Ensemble Empirical Mode Decomposition: A noise-assisted data analysis method, Adv. Adapt. Data Anal., 1, 1-41.

Wu, Z., N.-E. Huang, J.-M. Wallace, et al., 2011: On the time-varying trend in global-mean surface temperature, Climate Dyn., 37, 759-773.

Xia, J.-J., K. Tu, Z.-W. Yan, et al., 2015: The super heatwave in eastern China during July-August 2013: A perspective of climate change, Int. J. Climatol., doi:10.1002/joc.4424.

Xia, J.-J., and Z.-W. Yan, 2014: Changes in the local growing season in eastern China during 1909-2012, SOLA, 10, 163-166.

Xia, J.-J., Z.-W. Yan, and P.-L. Wu, 2013: Multidecadal variability in local growing season during 1901-2009, Climate Dyn., 41, 295-305.

Xu, M.-Z., and G.-Y. Ren, 2004: Change in growing season over China: 1961-2000, J. Appl. Meteor. Sci. (in Chinese), 15(3), 306-312.

Yan, Z.-W., C. Yang, and P.-D. Jones, 2001: Influence of Inhomogeneity on the estimation of mean and extreme temperature trends in Beijing and Shanghai, Adv. Atmos. Sci., 18(3), 309-322.

Yang, X.-C., Y.-L. Hou, and B.-D. Chen, 2011: Observed surface warming induced by urbanization in east China, J. Geophys. Res., 116, D14113, doi:10.1029/2010JD015452.

Yang, X.-C., Z. Tian, and B. Chen, 2013: Thermal growing season trends in east China, with emphasis on urbanization effects, Int. J. Climatol., 33, 2402-2412.

Zhao, P., P.-D. Jones, L.-J. Cao, et al., 2014: Trend of surface air temperature in eastern China and associated large-scale climate variability over the last 100 years, J. Climate, 27, 4693-4703.

Xia, J.-J., Z.-W. Yan, and Z. Xiong, 2015: Changes in local growing season indices in Shanghai due to urbanization during 1873-2013, Atmos. Oceanic Sci. Lett., 8, 383-385,

10.3878/AOSL20150044.

29 April 2015; revised 25 June 2015; accepted 6 July 2015; published 16 November 2015

XIONG Zhe, xzh@tea.ac.cn