CSIR-UGC NET — Earth, Atmospheric, Ocean and Planetary Sciences · India · National Testing Agency for CSIR
Original practice questions written to the published syllabus · not past papersHow future climate is projected and how to use projections responsibly: model hierarchy, CMIP ensembles, RCP and SSP scenarios, downscaling and bias correction, and uncertainty. Needed for impact studies, climate-risk disclosure and adaptation planning.
In the exams: GARP SCR 'Climate Models and Scenario Analysis'; PG and PhD research in agrometeorology and climate change; ICAR AICE subject 60 (climate variability)
CSIR-UGC NET — Earth, Atmospheric, Ocean and Planetary Sciences
Energy-balance models and response time, GCM structure and parameterisation, predictability and chaos, ENSO and MJO, forecast verification and scenario frameworks.
How our questions for CSIR-UGC NET spread across the units of Climate Models, Scenarios & Downscaling. Difficulty: 11 easy, 30 medium, 19 hard.
| Statistical and dynamical downscaling | 8 | |
| Bias correction methods | 8 | |
| Seasonal and sub-seasonal forecasting | 8 | |
| Climate indices and extreme-event indices | 8 | |
| Model hierarchy: energy-balance, GCM, ESM and regional models | 7 | |
| Using climate data portals and open datasets | 7 | |
| CMIP ensembles and model evaluation | 6 | |
| Sources of uncertainty: scenario, model and internal variability | 5 | |
| Representative Concentration Pathways and Shared Socioeconomic Pathways | 1 | |
| Integrated assessment models and mitigation pathways | 1 | |
| Scenario analysis for organisations: physical and transition scenarios | 1 |
Choose an answer, then open the working. Each explanation says why the right answer is right and why the others are not.
Why do climate models need parameterisations?
B. Sub-grid processes such as convective clouds cannot be resolved explicitly
Sub-grid processes such as convection, cloud microphysics and turbulence are represented by parameterisations in terms of resolved variables. They are not legal requirements, the large-scale flow is resolved from physical laws, and models are still evaluated against observations.
Further reading: See the manual sources: IPCC AR6 WGI/WGIII, CMIP6 design papers, ETCCDI and the documents named in the explanation
A consultant presents a 1 km downscaled projection from one global model as precise local guidance. What is the main problem?
C. Finer grids add detail, not accuracy, and one model hides most of the uncertainty
High resolution can give false precision; the projection inherits the driving model's large-scale behaviour, and one model ignores model uncertainty. 1 km products exist, single models are not best by default, and downscaling does not remove all bias or make forecasts.
Further reading: See the manual sources: IPCC AR6 WGI/WGIII, CMIP6 design papers, ETCCDI and the documents named in the explanation
Using linear scaling, a model's reference-period mean July maximum temperature is 28.5 °C and the observed mean is 27.0 °C. The model projects 29.4 °C for a future July. What is the corrected value?
B. 27.9 °C
Additive correction: 29.4 + (27.0 − 28.5) = 27.9 °C. 30.9 applies the bias with the wrong sign; 27.0 is the observed mean, which ignores the projected change; 29.4 is uncorrected.
Further reading: See the manual sources: IPCC AR6 WGI/WGIII, CMIP6 design papers, ETCCDI and the documents named in the explanation
For rainfall, the reference-period observed mean is 4.2 mm day⁻¹ and the model's is 3.0 mm day⁻¹. Using multiplicative linear scaling, what is the corrected value of a future model day with 3.6 mm?
A. 5.04 mm
Corrected = 3.6 × (4.2/3.0) = 5.04 mm. Adding the difference gives 4.80 (additive scaling, which can produce negative rain on dry days); inverting the ratio gives 2.57; 3.60 is uncorrected.
Further reading: See the manual sources: IPCC AR6 WGI/WGIII, CMIP6 design papers, ETCCDI and the documents named in the explanation
Why is precipitation usually bias-corrected multiplicatively rather than additively?
A. Multiplying keeps rain non-negative; adding a negative bias gives negative rain
A ratio correction keeps zero as zero and avoids negative values; additive correction can push dry days below zero. Rainfall is not measured in kelvin, there is no ban, and ratio scaling corrects only the mean.
Further reading: See the manual sources: IPCC AR6 WGI/WGIII, CMIP6 design papers, ETCCDI and the documents named in the explanation
Roughly how long does deterministic predictability of day-to-day weather last before chaos destroys skill?
A. About two weeks
Lorenz's work on chaos implies deterministic skill is lost after roughly two weeks; beyond that, skill comes from slowly varying boundary conditions and is probabilistic.
Further reading: See the manual sources: IPCC AR6 WGI/WGIII, CMIP6 design papers, ETCCDI and the documents named in the explanation
Four forecasts give probabilities of 0.8, 0.6, 0.3 and 0.1 for heavy rain; the event occurred on the first and third occasions only. What is the Brier score?
D. 0.225
BS = [(0.8−1)² + (0.6−0)² + (0.3−1)² + (0.1−0)²]/4 = (0.04 + 0.36 + 0.49 + 0.01)/4 = 0.225. 0.400 uses absolute errors; 0.900 forgets to divide by 4; 0.100 is the Brier skill score against a 0.5 climatology.
Further reading: See the manual sources: IPCC AR6 WGI/WGIII, CMIP6 design papers, ETCCDI and the documents named in the explanation
How is the Dipole Mode Index for the Indian Ocean Dipole defined?
B. West minus south-east tropical Indian Ocean SST anomaly
The DMI is the SST anomaly difference between the western and south-eastern equatorial Indian Ocean. Tahiti–Darwin is the Southern Oscillation Index, Niño 3.4 is ENSO, and Azores–Iceland is the NAO.
Further reading: See the manual sources: IPCC AR6 WGI/WGIII, CMIP6 design papers, ETCCDI and the documents named in the explanation
ERA5, from ECMWF and the Copernicus Climate Change Service, provides
B. hourly global reanalysis at 31 km from 1940 onwards
ERA5 is an hourly reanalysis at ~0.25° (31 km) from 1940. CHIRPS is the 0.05° rainfall product; the other descriptions do not fit ERA5.
Further reading: See the manual sources: IPCC AR6 WGI/WGIII, CMIP6 design papers, ETCCDI and the documents named in the explanation
Where is raw CMIP6 model output primarily distributed?
A. The Earth System Grid Federation and the Copernicus Data Store
CMIP data are served through ESGF nodes, and many variables via the Copernicus CDS. UNFCCC portals hold inventories, and statistics offices do not distribute CMIP output.
Further reading: See the manual sources: IPCC AR6 WGI/WGIII, CMIP6 design papers, ETCCDI and the documents named in the explanation
Ten here, 60 in the app. The Laisha app has every practice question we have written to CSIR-UGC NET in this subject, marked as you go, with timed papers, a jump-to-unanswered list and a certificate when you pass at 70 per cent.
Free for ten days, then one subscription of US$10 a month. We would rather say so here than have you find it later.
An AI operating system for agriculture: free guides, manuals and tools written for every country.