INFLUENCE OF LAND SURFACE PROCESSES ON SIMULATION OF WESTERN DISTURBANCES AND
ASSOCIATED PRECIPITATION
by
LIBY THOMAS
Centre for Atmospheric Sciences
Submitted
in fulfillment of the requirements of the degree of Doctor of Philosophy
to the
INDIAN INSTITUTE OF TECHNOLOGY DELHI
MAY 2013
Dedicated to My Parents
Certificate
This is to certify that the thesis entitled “INFLUENCE OF LAND SURFACE PROCESSES ON SIMULATION OF WESTERN DISTURBANCES AND ASSOCIATED PRECIITATION” being submitted by Mr. Liby Thomas for the award of the degree of DOCTOR OF PHYLOSOPHY is a record of the original bonafide research work carried out by him under our joint guidance and supervision at the Centre for Atmospheric Sciences, Indian Institute of Technology, Delhi. The work presented in this thesis has not been submitted elsewhere either in part or full to any other University or Institute for award of any degree or diploma.
Professor U. C. Mohanty Professor S. K. Dash
Centre for Atmospheric Sciences Centre for Atmospheric Sciences Indian Institute of Technology Delhi Indian Institute of Technology Delhi Hauz Khas, New Delhi-110016, India Hauz Khas, New Delhi-110016, India
New Delhi May 2013
Acknowledgements
I was blessed with many wonderful teachers and I am deeply indebted to them all.
It is my greatest pleasure and honor to sincerely thank Prof. U.C. Mohanty and Prof. S.K.
Dash for their valuable guidance, constructive criticism, constant encouragement and ceaseless help extended throughout this research work. They were my ‘steadfast’ well wishers and moral support in many difficult times at IIT Delhi.
I am also grateful to Prof. Maithili Sharan and Prof. O.P. Sharma, former heads, Centre for Atmospheric Sciences for their immense help and encouragement during the course of this work. My sincere thanks are due to all the faculty members of CAS for their suggestions and support during the research. I also wish to extend my deep appreciation to the SRC members Prof. A. D. Rao and Prof. G. N. Tiwari, Center for Energy Studies, IIT Delhi.
I sincerely acknowledge National Centre for Atmospheric Research (NCAR), USA and National Centers for Environmental Prediction (NCEP), USA for providing the model and reanalysis data sets. European Centre for Medium-Range Weather Forecasts (ECMWF) and National Remote Sensing Centre (NRSC) are also acknowledged for providing ERA 40 data and AWiFS land use data. India Meteorological Department (IMD) is highly appreciated and acknowledged for providing the rainfall and temperature data sets and weather reports.
I gratefully acknowledge the financial support of Council of Scientific and Industrial Research (CSIR) during this research work. My sincere thanks are also due to all supporting staff members of CAS for their all sorts of help, in particular, L. S. Negi, V. K. Kaushik, Krishnan Kumar, Ajay Kumar Sethi, Data Ram, Kedar Nath, Mrs. Kusum Sehrawat, Mrs. Saroj Bala Gupta, S. S. Negi and Prasanth. I express my thanks to all my colleagues and friends at CAS, IIT Delhi, in particular, Narender, Dr. Ashish, Dr. Palash, Dr. Sujata, Mourani, Krishna, Kiran, Litta, Rajesh, Rajeevan, Dr. Sankalp, Dr. Subrat, Rajeev, Dileep, Himansu, Dr. Pramod, Dr. Senthil, Suraj Singh Saini, Dipak, Kanhu, Srinivas, Anikender, Dr. Sarvesh, Rani, Lalit, Dr. Swagata, Gaurav, Pushparaj, Dhiru,
Sathiyaseelan, Abhishek and Bhuvan for sharing companionship and making the stay at IIT pleasant.
I thank Manu and Suraj Harshan for his immense help in the early days at IIT Delhi. I wish to thank all my friends in particular Dr. Anith, Dr. Jay, Girish, Dr. Sabin, Dr. Rakesh, Dr. Dipu, Dr. Vijeesh, Tom, Bibin, Praju and Priya for their help and encouragement. I express my sincere thanks to a large number of school and college day friends for their concern and good wishes.
I thank my family for their support and encouragement. I thank my parents, Shri.
Thomas and Smt. Lissy Thomas who has been the invaluable boon of Lord for their moral support, great understanding, funds, advice, and help all throughout my research work. I dedicate this thesis to them.
New Delhi Liby Thomas
Abstract
The northern states of India such as Jammu and Kashmir, Punjab, Himachal Pradesh, Haryana, Uttarakhand, Northern Rajasthan and Northwest Uttar Pradesh primarily get rainfall during the winter months from the Western Disturbances (WDs).
WDs are eastward moving low pressure synoptic weather systems, which originate over Mediterranean Sea or mid Atlantic Ocean and travel towards east over Iran, Afghanistan, Pakistan and north India. The WDs significantly affect weather over Iran, Iraq, Afghanistan, Pakistan and India, especially during the post monsoon and winter months from October to March and these disturbances also yield enormous amount of precipitation. These weather systems take southernmost tracks during winter season and pass through north Indian region. The strength and vertical extent of these systems are influenced by position of the mid and upper tropospheric troughs in the zonal westerlies.
Further, northern parts of India comprises of complex mountain ranges and hence variable terrain and complex orography. Due to highly variable altitude and orientation of orographic barriers, the prevailing weather conditions over the region are very complex.
Amount of winter precipitation received during December to March (DJFM) depends on frequency and intensity of WDs. Not only the physical processes such as convection, release of latent heat but also the regional topographical features play important role in the evolution of WD. Paucity of adequate data has made numerical weather prediction a very useful tool to study WDs in the region. In fact there are a very limited number of numerical studies over the study region.
The primary objective of the thesis is to understand the influence of various land surface parameters for the simulation of WDs using Weather Research and Forecasting
(WRF) model. The thesis is divided into six chapters. Chapter one is an introductory chapter about the weather features over north India during the passage of WDs and importance of land surface parameters. It also includes the literature review done on WDs. Further, the objective and scope of the thesis is mentioned in this chapter.
Circulation features and precipitation associated with WDs over north India has been analyzed in the second chapter. The primary objective of this chapter is to examine ERA 40 data set to compare the characteristics of winter season over the north India as seen in the IMD rainfall and temperature gridded data sets. The amount and spatial distribution of rainfall from both ERA 40 data and IMD rainfall data for the winter months over north India are found to be very much similar. By using 30 years (1971- 2001) of past data it is found that there exists a strong positive correlation between the two data sets except the extreme northeast part of north India during the winter months (December, January, February and March) and hence ERA 40 data are used here for studying the winter time circulation characteristics over this region. At most of the grid points the range of correlation coefficient varies from 0.4 to 0.9 for rainfall except at a few points in the northeastern part of India. For temperature, the correlation coefficient varies from 0.5 to 0.95. Winter circulation features from ERA 40 data are also analyzed.
One of the major rain producing weather phenomena of north India during winter months belongs to the category of WDs. NCEP-DOE reanalysis 2 data and Tropical Rainfall Measuring Mission (TRMM) data are used to examine the circulation features and precipitation during the occurrence of WDs over north India. Cyclonic circulations at 850 hPa, troughs and ridges at 500 hPa and westerly winds at 200 hPa are obtained from the
NCEP-DOE reanalysis 2 data. It is clear from TRMM data sets that widespread rainfall is observed in north India due to the passage of WDs.
Eight WDs are simulated using Advanced Research WRF (ARW) model with different land surface parameterization (LSP) schemes in third chapter. The LSP schemes used in this study are thermal diffusion, Noah and rapid update cycle (RUC). The model simulated circulation features and rainfall amounts are compared with respective National Centre for Environmental Prediction (NCEP) Final Analysis (FNL) data sets and TRMM observed rainfall in order to select the best LSP scheme. The model simulated rainfall is also compared with those obtained from India Meteorological Department (IMD) station observations. Results indicate that circulation features and precipitation simulated by the model are sensitive to the selection of LSP schemes. Rainfall is slightly overpredicted by the model in most of the cases. The rainfall averaged over the region encompassing 72oE- 83oE to 27.5oN-38.5oN obtained from TRMM data sets are compared with corresponding model simulations. Comparison shows that the spatial distribution and the area averaged rainfall simulated by RUC scheme is close to that obtained from TRMM. Further, the root mean square error (RMSE) values of area averaged rainfall using three LSP schemes from the eight cases are calculated against corresponding TRMM data. The results show that thermal diffusion has the RMSE value of 0.54, Noah LSP scheme has 0.45 and RUC LSP scheme has the least value of 0.23. The CC between IMD observed rainfall at various stations and corresponding model simulated values obtained using three LSP schemes are also compared. Results show that RUC LSP scheme yields higher value of CC at various observing stations compared to those obtained in other schemes. The simulated relative humidity, mean sea level pressure, winds at 850 hPa, 500 hPa and 200
hPa are compared with respective FNL fields. The patterns of relative humidity and mean sea level pressure are captured relatively well in model simulations when RUC scheme is used. The wind magnitudes are found to be stronger compared with the corresponding FNL analysis.
The sensitivity of soil moisture to the simulation of precipitation associated with eight WDs are investigated using ARW model in the fourth chapter. For this purpose one control experiment is done for each of eight cases of WDs with the best suitable combination of parameterization schemes (from chapter 3). Then the simulations are repeated with 10 % increase and 10 % decrease in the soil moisture respectively. The area averaged rainfall over northern India from model simulations is compared with corresponding TRMM observations. Results show that there is 1.7 % increase in the precipitation when there is 10% increase in soil moisture. Similarly there is 1.4 % reduction in precipitation when there is 10 % decrease in soil moisture. Model simulated values of rainfall are also compared with corresponding IMD observed values at different stations. Rainfall corresponding to station observations is enhanced (reduced) when soil moisture is increased (decreased).
In chapter five, the new land use data over India obtained from Advanced Wide Field Sensor (AWiFS) aboard the Indian satellite IRS P6 in the ARW model are used to investigate their effects on precipitation and circulation features in the simulation of three WDs during 2008 and 2009. A detailed intercomparison of the model’s response to the AWiFS data as against USGS data is made in this chapter. It is found that AWiFS data over India during 2008 and 2009 are very different from USGS data generated during April 1992 to March 1993 based on NOAA AVHRR images. The AWiFS data represents
the recent changes in the land use over India and hence more realistic than USGS data.
The model simulated circulation features are compared with corresponding NCEP FNL fields and further rainfall values are compared with TRMM and IMD observations.
Model simulations indicate that the rainfall and circulation parameters simulated by the model are improved when AWiFS data are used in comparison to the USGS data.
Chapter six gives brief summary drawn from the studies described in the earlier chapters. Chapter one give an introduction about the weather features over north India during the passage of WDs, importance of land surface parameters and scope and objectives of the thesis. Circulation features and precipitation associated with WDs over north India has been analyzed in the second chapter. In the third chapter sensitivity study is carried out to find the suitable land surface scheme for the simulation of WDs and it is found that simulation using RUC land surface scheme give better rainfall and circulation features compared to the other schemes. In the fourth chapter the sensitivity of soil moisture to the simulation of precipitation associated with WDs are investigated using ARW model and found that rainfall amounts increase (decrease) by increasing (decreasing) soil moisture. Comparison of simulations using AWiFS land use data and USGS land use data is carried out in the fifth chapter. Simulations with AWiFS data produced better rainfall and circulation features compared to USGS data.
Numerical simulations give better results when the horizontal resolution is increased. Hence, in the future studies high resolution nested domain will be used for simulation of WDs. The numerical experiments will be carried out to improve the model performance through mesoscale data assimilation technique (3DVAR/4DVAR) using
available quality controlled observations. Additional improvements could be obtained using the land surface data assimilation system (LDAS) in the ARW model
Contents
Certificate
Acknowledgements Abstract
Contents List of Figures List of Tables
Chapter 1 Introduction……….…1
1.1 Western Disturbances………3
1.2 Importance of land surface parameters……….9
1.3 Numerical Weather Prediction……….…14
1.3.1 General Assumptions in Numerical Modelling………16
1.3.2 Parameterization of Atmospheric Processes……….……21
1.4 Objectives and Scope of Thesis………...23
Chapter 2 Characteristics of Western Disturbances over north India as revealed in reanalysed data sets...26
2.1 Synoptic situations during selected WDs………29
2.2 Data sets………...……32
2.3 Characteristics of winter circulation and rainfall over north India……….……36
2.3.1 Rainfall………..36
2.3.2 Temperature………..41
2.3.3 Wind and mean sea level pressure………43
2.4 Rainfall and circulation features associated with WDs………...…47
2.5 Summary………..………53
Chapter 3 Influence of various land surface parameterization schemes on the simulation of Western Disturbances.………..……..55
3.1 Description of Weather Research and Forecasting (WRF) modelling system………57
3.1.1 Features of the ARW System………58
3.1.2 The Governing Equations……….61
3.2 LSP schemes………66
3.3 Synoptic situation during selected Western Disturbances………..…68
3.4 Experimental design……….…70
3.5 Characteristics of Western Disturbances simulated by ARW model………..…73
3.5.1 Rainfall………..73
3.5.2 Circulation features………...…86
3.5.3 Relative humidity………..…88
3.5.4 Mean sea level pressure………89
3.6. Summary……….…90
Chapter 4 Impact of soil moisture on the simulation of WDs and associated precipitation ………....92
4.1 Synoptic situation during selected Western Disturbances………...96
4.2 Experimental design………98
4.3 Cases studied………99
4.4 Important features of Western Disturbances simulated by ARW model………...…100
4.4.1 Rainfall………100
4.4.2 Circulation features……….…109
4.4.3 Mean sea level pressure………..111
4.4.4 Relative Humidity………...112
4.5 Summary………113
Chapter 5 Impact of different land use on WD simulations………..……115
5.1 Characteristics of AWiFS land use data………118
5.2 Synoptic situation during selected Western Disturbances……….…122
5.3 Experimental design………...…125
5.4 Results and discussions………..126
5.4.1 Rainfall………126
5.4.2 Circulation features……….…134
5.4.3 Relative humidity and mean sea level pressure………..…136
5.5 Summary………138
Chapter 6 Conclusions………...140
References………...147
Bio-Data………..171
