Application of a bivariate bias-correction approach to yield long-term attributes of Indian precipitation and temperature
The General Circulation Model (GCM) simulation had shown potential in yielding long-term statistical attributes of Indian precipitation and temperature which exhibit substantial inter-seasonal variation. However, GCM outputs experience substantial model structural bias that needs to be reduced prior...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2023-05-01
|
| Series: | Frontiers in Climate |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fclim.2023.1067960/full |
| _version_ | 1797833049320194048 |
|---|---|
| author | Chanchal Gupta Rajarshi Das Bhowmik |
| author_facet | Chanchal Gupta Rajarshi Das Bhowmik |
| author_sort | Chanchal Gupta |
| collection | DOAJ |
| description | The General Circulation Model (GCM) simulation had shown potential in yielding long-term statistical attributes of Indian precipitation and temperature which exhibit substantial inter-seasonal variation. However, GCM outputs experience substantial model structural bias that needs to be reduced prior to forcing them into hydrological models and using them in deriving insights on the impact of climate change. Traditionally, univariate bias correction approaches that can successfully yield the mean and the standard deviation of the observed variable, while ignoring the interdependence between multiple variables, are considered. Limited efforts have been made to develop bivariate bias-correction over a large region with an additional focus on the cross-correlation between two variables. Considering these, the current study suggests two objectives: (i) To apply a bivariate bias correction approach based on bivariate ranking to reduce bias in GCM historical simulation over India, (ii) To explore the potential of the proposed approach in yielding inter-seasonal variations in precipitation and temperature while also yielding the cross-correlation. This study considers three GCMs with fourteen ensemble members from the Coupled Model Intercomparison project Assessment Report-5 (CMIP5). The bivariate ranks of meteorological pairs are applied on marginal ranks till a stationary position is achieved. Results show that the bivariate approach substantially reduces bias in the mean and the standard deviation. Further, the bivariate approach performs better during non-monsoon months as compared to monsoon months in reducing the bias in the cross-correlation between precipitation and temperature as the typical negative cross-correlation structure is common during non-monsoon months. The study finds that the proposed approach successfully reproduces inter-seasonal variation in metrological variables across India. |
| first_indexed | 2024-04-09T14:17:28Z |
| format | Article |
| id | doaj.art-8254eec6e7dc495682885cc2f97a96d3 |
| institution | Directory Open Access Journal |
| issn | 2624-9553 |
| language | English |
| last_indexed | 2024-04-09T14:17:28Z |
| publishDate | 2023-05-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Climate |
| spelling | doaj.art-8254eec6e7dc495682885cc2f97a96d32023-05-05T06:08:25ZengFrontiers Media S.A.Frontiers in Climate2624-95532023-05-01510.3389/fclim.2023.10679601067960Application of a bivariate bias-correction approach to yield long-term attributes of Indian precipitation and temperatureChanchal GuptaRajarshi Das BhowmikThe General Circulation Model (GCM) simulation had shown potential in yielding long-term statistical attributes of Indian precipitation and temperature which exhibit substantial inter-seasonal variation. However, GCM outputs experience substantial model structural bias that needs to be reduced prior to forcing them into hydrological models and using them in deriving insights on the impact of climate change. Traditionally, univariate bias correction approaches that can successfully yield the mean and the standard deviation of the observed variable, while ignoring the interdependence between multiple variables, are considered. Limited efforts have been made to develop bivariate bias-correction over a large region with an additional focus on the cross-correlation between two variables. Considering these, the current study suggests two objectives: (i) To apply a bivariate bias correction approach based on bivariate ranking to reduce bias in GCM historical simulation over India, (ii) To explore the potential of the proposed approach in yielding inter-seasonal variations in precipitation and temperature while also yielding the cross-correlation. This study considers three GCMs with fourteen ensemble members from the Coupled Model Intercomparison project Assessment Report-5 (CMIP5). The bivariate ranks of meteorological pairs are applied on marginal ranks till a stationary position is achieved. Results show that the bivariate approach substantially reduces bias in the mean and the standard deviation. Further, the bivariate approach performs better during non-monsoon months as compared to monsoon months in reducing the bias in the cross-correlation between precipitation and temperature as the typical negative cross-correlation structure is common during non-monsoon months. The study finds that the proposed approach successfully reproduces inter-seasonal variation in metrological variables across India.https://www.frontiersin.org/articles/10.3389/fclim.2023.1067960/fullbias-correctionBABCGCM (General Circulation Model)monsooncorrelationACCA |
| spellingShingle | Chanchal Gupta Rajarshi Das Bhowmik Application of a bivariate bias-correction approach to yield long-term attributes of Indian precipitation and temperature Frontiers in Climate bias-correction BABC GCM (General Circulation Model) monsoon correlation ACCA |
| title | Application of a bivariate bias-correction approach to yield long-term attributes of Indian precipitation and temperature |
| title_full | Application of a bivariate bias-correction approach to yield long-term attributes of Indian precipitation and temperature |
| title_fullStr | Application of a bivariate bias-correction approach to yield long-term attributes of Indian precipitation and temperature |
| title_full_unstemmed | Application of a bivariate bias-correction approach to yield long-term attributes of Indian precipitation and temperature |
| title_short | Application of a bivariate bias-correction approach to yield long-term attributes of Indian precipitation and temperature |
| title_sort | application of a bivariate bias correction approach to yield long term attributes of indian precipitation and temperature |
| topic | bias-correction BABC GCM (General Circulation Model) monsoon correlation ACCA |
| url | https://www.frontiersin.org/articles/10.3389/fclim.2023.1067960/full |
| work_keys_str_mv | AT chanchalgupta applicationofabivariatebiascorrectionapproachtoyieldlongtermattributesofindianprecipitationandtemperature AT rajarshidasbhowmik applicationofabivariatebiascorrectionapproachtoyieldlongtermattributesofindianprecipitationandtemperature |