Comparison and correction of IDW based wind speed interpolation methods in urbanized Shenzhen

doi:10.1007/s11707-021-0948-z

Front. Earth Sci.

2022, Vol. 16

Issue (3) : 798-808 https://doi.org/10.1007/s11707-021-0948-z

RESEARCH ARTICLE

Comparison and correction of IDW based wind speed interpolation methods in urbanized Shenzhen

Wei ZHAO¹, Yuping ZHONG¹, Qinglan LI¹(

), Minghua LI², Jia LIU², Li TANG²

¹. Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, Shenzhen 518055, China
². Shenzhen Meteorological Bureau, Shenzhen 518040, China

Download: PDF(3664 KB) HTML
Export: BibTeX | EndNote | Reference Manager | ProCite | RefWorks

Abstract

Based on the 2-min average wind speed observations at 100 automatic weather stations in Shenzhen from January 2008 to December 2018, this study tries to explore the ways to improve wind interpolation quality over the Shenzhen region. Three IDW based methods, i.e., traditional inverse distance weight (IDW), modified inverse distance weight (MIDW), and gradient inverse distance weight (GIDW) are used to interpolate the near surface wind field in Shenzhen. In addition, the gradient boosted regression trees (GBRT) model is used to correct the wind interpolation results based on the three IDW based methods. The results show that among the three methods, GIDW has better interpolation effects than the other two in the case of stratified sampling. The MSE and R² for the GIDW’s in different months are in the range of 1.096–1.605 m/s and 0.340–0.419, respectively. However, in the case of leave-one-group-out cross-validation, GIDW has no advantage over the other two methods. For the stratified sampling, GBRT effectively corrects the interpolated results by the three IDW based methods. MSE decreases to the range of 0.778–0.923 m/s, and R² increases to the range of 0.530–0.671. In the non-station area, the correction effect of GBRT is still robust, even though the elevation frequency distribution of the non-station area is different from that of the stations’ area. The correction performance of GBRT mainly comes from its consideration of the nonlinear relationship between wind speed and the elevation, and the combination of historical and current observation data.

Keywords wind interpolation Shenzhen inverse distance weight gradient boosted regression trees

Corresponding Author(s): Qinglan LI

Online First Date: 30 June 2022 Issue Date: 29 December 2022

Cite this article:

Wei ZHAO,Yuping ZHONG,Qinglan LI, et al. Comparison and correction of IDW based wind speed interpolation methods in urbanized Shenzhen[J]. Front. Earth Sci., 2022, 16(3): 798-808.

URL:

https://academic.hep.com.cn/fesci/EN/10.1007/s11707-021-0948-z
https://academic.hep.com.cn/fesci/EN/Y2022/V16/I3/798

Fig.1 Location of the 100 automatic weather stations (AWSs) and the geographical features of the study area.

Tab.1 Input features and regression target of GBRT

Month	MSE			$R 2$
Month	IDW	MIDW	GIDW	IDW	MIDW	GIDW
Jan.	1.487	1.514	1.440	0.351	0.339	0.371
Feb.	1.425	1.459	1.393	0.364	0.349	0.379
Mar.	1.364	1.375	1.302	0.364	0.359	0.393
Apr.	1.307	1.332	1.264	0.345	0.332	0.366
May	1.265	1.309	1.240	0.327	0.303	0.340
Jun.	1.271	1.350	1.267	0.358	0.319	0.361
Jul.	1.241	1.310	1.229	0.355	0.319	0.361
Aug.	1.081	1.137	1.096	0.411	0.381	0.404
Sep.	1.278	1.319	1.254	0.408	0.389	0.419
Oct.	1.368	1.410	1.334	0.347	0.327	0.363
Nov.	1.464	1.496	1.406	0.366	0.352	0.391
Dec.	1.664	1.700	1.605	0.389	0.376	0.411
All year	1.350	1.392	1.318	0.368	0.348	0.383

Tab.2 Interpolation evaluation of IDW, MIDW, and GIDW

Month	MSE			$R 2$
Month	GBRIDW	GBRMIDW	GBRGIDW	GBRIDW	GBRMIDW	GBRGIDW
Jan.	0.880	0.878	0.878	0.615	0.617	0.617
Feb.	0.885	0.879	0.880	0.606	0.608	0.608
Mar.	0.881	0.874	0.877	0.589	0.592	0.591
Apr.	0.891	0.889	0.890	0.553	0.554	0.553
May	0.883	0.882	0.883	0.530	0.531	0.530
Jun.	0.922	0.920	0.923	0.535	0.536	0.534
Jul.	0.881	0.880	0.882	0.542	0.543	0.541
Aug.	0.779	0.778	0.780	0.576	0.577	0.576
Sep.	0.844	0.841	0.840	0.609	0.610	0.611
Oct.	0.837	0.836	0.835	0.600	0.601	0.601
Nov.	0.835	0.830	0.831	0.638	0.641	0.640
Dec.	0.901	0.897	0.899	0.669	0.671	0.670
All year	0.868	0.865	0.866	0.594	0.596	0.595

Tab.3 Performance evaluation of the GBRT correction

Fig.2 The R² and MSE of each station’s interpolation result for the testing set by IDW, MIDW, and GIDW methods. (a) (c) (e) for R², and (b) (d) (f) for MSE.

Fig.3 Interpolation improvement by GBRIDW, GBRMIDW, and GBRGIDW methods in terms of ?R² and ?MSE, comparing to the interpolations by IDW, MIDW, and GIDW methods.

Fig.4 Comparison of (a) elevation and (b) slope distribution between the 100 stations and the natural terrain in Shenzhen.

Fig.5 The locations of the stations in the three randomly selected groups.

Group	MSE			$R 2$
Group	IDW	MIDW	GIDW	IDW	MIDW	GIDW
1	1.223	1.297	1.471	0.321	0.280	0.183
2	1.392	1.598	1.631	0.326	0.227	0.211
3	1.193	1.313	1.497	0.364	0.300	0.202

Tab.4 Interpolation evaluation of IDW, MIDW, and GIDW

Group	MSE			$R 2$
Group	GBR-IDW	GBRM-IDW	GBRG-IDW	GBR-IDW	GBRM-IDW	GBRG-IDW
1	1.210	1.181	1.288	0.328	0.344	0.285
2	1.356	1.384	1.453	0.344	0.330	0.297
3	1.190	1.214	1.306	0.366	0.353	0.304

Tab.5 Performance evaluation of the GBRT correction

Fig.6 The importance of each input feature. (a) and (b) are for GBRIDW; (c) and (d) are for GBRMIDW; (e) and (f) are for GBRGIDW.

1	F J Aguilar, F Agüera, M A Aguilar, F Carvajal. ( 2005). Effects of terrain morphology, sampling density, and interpolation methods on grid DEM accuracy. Photogramm Eng Remote Sensing, 71( 7): 805– 816 https://doi.org/10.14358/PERS.71.7.805
2	A Altmann, L Toloşi, O Sander, T Lengauer. ( 2010). Permutation importance: a corrected feature importance measure. Bioinformatics, 26( 10): 1340– 1347 https://doi.org/10.1093/bioinformatics/btq134
3	M A Arain, R Blair, N Finkelstein, J R Brook, T Sahsuvaroglu, B Beckerman, L Zhang, M Jerrett. ( 2007). The use of wind fields in a land use regression model to predict air pollution concentrations for health exposure studies. Atmos Environ, 41( 16): 3453– 3464 https://doi.org/10.1016/j.atmosenv.2006.11.063
4	F Bañuelos-Ruedas, C Angeles-Camacho, S Rios-Marcuello. ( 2010). Analysis and validation of the methodology used in the extrapolation of wind speed data at different heights. Renew Sustain Energy Rev, 14( 8): 2383– 2391 https://doi.org/10.1016/j.rser.2010.05.001
5	G Casalicchio C Molnar B Bischl ( 2018). Visualizing the feature importance for black box models. In: Joint European Conference on Machine Learning and Knowledge Discovery in Databases. Cham: Springer
6	Q Chang, S Li, Y Wang, J Wu, M Xie. ( 2013). Spatial process of green infrastructure changes associated with rapid urbanization in Shenzhen, China. Chin Geogr Sci, 23( 1): 113– 128 https://doi.org/10.1007/s11769-012-0568-3
7	S Coccolo, D Mauree, E Naboni, J Kaempf, J L Scartezzini. ( 2017). On the impact of the wind speed on the outdoor human comfort: a sensitivity analysis. Energy Procedia, 122: 481– 486 https://doi.org/10.1016/j.egypro.2017.07.297
8	C L Curry, D van der Kamp, A H Monahan. ( 2012). Statistical downscaling of historical monthly mean winds over a coastal region of complex terrain. I. Predicting wind speed. Clim Dyn, 38( 7−8): 1281– 1299 https://doi.org/10.1007/s00382-011-1173-3
9	C Daly, M Halbleib, J I Smith, W P Gibson, M K Doggett, G H Taylor, J Curtis, P P Pasteris. ( 2008). Physiographically sensitive mapping of climatological temperature and precipitation across the conterminous United States. Intern J Climat, 28( 15): 2031– 2064 https://doi.org/10.1002/joc.1688
10	J Elith, J R Leathwick, T Hastie. ( 2008). A working guide to boosted regression trees. J Anim Ecol, 77( 4): 802– 813 https://doi.org/10.1111/j.1365-2656.2008.01390.x
11	J H Friedman. ( 2001). Greedy function approximation: a gradient boosting machine. Ann Stat, 29( 5): 1189– 1232 https://doi.org/10.1214/aos/1013203451
12	J H Friedman. ( 2002). Stochastic gradient boosting. Comput Stat Data Anal, 38( 4): 367– 378 https://doi.org/10.1016/S0167-9473(01)00065-2
13	S P Ghosh. ( 1958). A note on stratified random sampling with multiple characters. Calcutta Statistical Association Bulletin, 8( 2−3): 81– 90 https://doi.org/10.1177/0008068319580204
14	A Jarvis H I Reuter A Nelson E Guevara ( 2008). SRTM 90m Digital Elevation Database v4. 1
15	H K Johnson. ( 1999). Simple expressions for correcting wind speed data for elevation. Coast Eng, 36( 3): 263– 269 https://doi.org/10.1016/S0378-3839(99)00016-2
16	H Kikumoto, R Ooka, H Sugawara, J Lim. ( 2017). Observational study of power-law approximation of wind profiles within an urban boundary layer for various wind conditions. J Wind Eng Ind Aerodyn, 164: 13– 21 https://doi.org/10.1016/j.jweia.2017.02.003
17	L Kleerekoper, M Van Esch, T B Salcedo. ( 2012). How to make a city climate-proof, addressing the urban heat island effect. Resour Conserv Recycling, 64: 30– 38 https://doi.org/10.1016/j.resconrec.2011.06.004
18	J Li, Y Tian, Y Zhu, T Zhou, J Li, K Ding, J Li. ( 2020). A multicenter random forest model for effective prognosis prediction in collaborative clinical research network. Artif Intell Med, 103: 101814 https://doi.org/10.1016/j.artmed.2020.101814
19	L Li. ( 2019). Geographically weighted machine learning and downscaling for high-resolution spatiotemporal estimations of wind speed. Remote Sens, 11( 11): 1378 https://doi.org/10.3390/rs11111378
20	L Li, J Qian, C Q Ou, Y X Zhou, C Guo, Y Guo. ( 2014). Spatial and temporal analysis of air pollution index and its timescale-dependent relationship with meteorological factors in Guangzhou, China, 2001–2011. Environ Pollut, 190: 75– 81 https://doi.org/10.1016/j.envpol.2014.03.020
21	C Liu, Q Li, W Zhao, Y Wang, R Ali, D Huang, X Lu, H Zheng, X Wei. ( 2020). Spatiotemporal characteristics of near-surface wind in Shenzhen. Sustainability, 12( 2): 739 https://doi.org/10.3390/su12020739
22	G Louppe, L Wehenkel, A Sutera, P Geurts. ( 2013). Understanding variable importances in forests of randomized trees. In: Proceedings of the 26th International Conference on Neural Information Processing Systems, 431– 439
23	N Masseran, A M Razali, K Ibrahim. ( 2012). An analysis of wind power density derived from several wind speed density functions: the regional assessment on wind power in Malaysia. Renew Sustain Energy Rev, 16( 8): 6476– 6487 https://doi.org/10.1016/j.rser.2012.03.073
24	M H McCutchan, D G Fox. ( 1986). Effect of elevation and aspect on wind, temperature and humidity. J Clim Appl Meteorol, 25( 12): 1996– 2013 https://doi.org/10.1175/1520-0450(1986)025<1996:EOEAAO>2.0.CO;2
25	L Mitas H Mitasova ( 1999). Spatial interpolation. In: Longley P A, Goodchild M F, Maguire D J, Rhind D W, eds. Geographical Information Systems: Principles, Techniques, Management and Applications. New York: Wiley
26	H Morris M Wardle M Mainelli ( 2020). The global financial centres index 28. Long Finan Financ Centr Future. 1− 69
27	I A Nalder, R W Wein. ( 1998). Spatial interpolation of climatic normals: test of a new method in the Canadian boreal forest. Agric Meteorol, 92( 4): 211– 225 https://doi.org/10.1016/S0168-1923(98)00102-6
28	E Ozelkan, G Chen, B B Ustundag. ( 2016). Spatial estimation of wind speed: a new integrative model using inverse distance weighting and power law. Int J Digit Earth, 9( 8): 733– 747 https://doi.org/10.1080/17538947.2015.1127437
29	J P Palutikof, P M Kelly, T D Davies, J A Halliday. ( 1987). Impacts of spatial and temporal windspeed variability on wind energy output. J Clim Appl Meteorol, 26( 9): 1124– 1133 https://doi.org/10.1175/1520-0450(1987)026<1124:IOSATW>2.0.CO;2
30	S Peng ( 2017). Optimized study on spatial interpolation methods for meteorological element. Geospat Inform, 15( 7): 86− 89 (in Chinese)
31	W R Tobler. ( 1970). A computer movie simulating urban growth in the Detroit region. Econom Geogra, 46( sup1): 234– 240
32	J Wang, J Hu, K Ma. ( 2016a). Wind speed probability distribution estimation and wind energy assessment. Renew Sustain Energy Rev, 60: 881– 899 https://doi.org/10.1016/j.rser.2016.01.057
33	P Wang, Y Dou, Y Xin. ( 2016b). The analysis and design of the job recommendation model based on GBRT and time factors. In: 2016 IEEE International Conference on Knowledge Engineering and Applications (ICKEA). IEEE, 29– 35
34	Y Wang Y (2019) Tang. A recommendation algorithm based on item genres preference and GBRT. In: Journal of Physics Conference Series. IOP Publishing, 2019, 1229( 1): 012053
35	R Wanninkhof. ( 2014). Relationship between wind speed and gas exchange over the ocean revisited. Limnol Oceanogr Methods, 12( 6): 351– 362 https://doi.org/10.4319/lom.2014.12.351
36	A Winstral, T Jonas, N Helbig. ( 2017). Statistical downscaling of gridded wind speed data using local topography. J Hydrometeorol, 18( 2): 335– 348 https://doi.org/10.1175/JHM-D-16-0054.1
37	Y Xu X Wan C Fu C Liu ( 2012). Wind speed interpolation under complex terrain conditions: a case study of Jilin Province. Yunnan Geogr Environ Res, 24 ( 4), 78− 81 (in Chinese)
38	J Yang, B Shi, Y Shi, S Marvin, Y Zheng, G Xia. ( 2020). Air pollution dispersal in high density urban areas: research on the triadic relation of wind, air pollution, and urban form. Sustain Cities Soc, 54: 101941 https://doi.org/10.1016/j.scs.2019.101941
39	W Yuan, B Xu, Z Chen, J Xia, W Xu, Y Chen, X Wu, Y Fu. ( 2015). Validation of China-wide interpolated daily climate variables from 1960 to 2011. Theor Appl Climatol, 119( 3−4): 689– 700 https://doi.org/10.1007/s00704-014-1140-0
40	Z Zhang, W Yang, S Wushour. ( 2020). Traffic accident prediction based on LSTM-GBRT model. J Contr Sci Eng, 2020: 4206919 https://doi.org/10.1155/2020/4206919
41	S Zhi, G Qian, J Luo. ( 2001). The variation of coastal wind with height in Guangdong Province. Trop Geogr, 21( 2): 131– 134

Viewed

Full text

Abstract

Cited

Shared

Discussed