Modelling the climate sensitivity of proso millet (panicum miliaceum L.) in Sri Lanka.

Wimalasiri, G. E. M. (2019) Modelling the climate sensitivity of proso millet (panicum miliaceum L.) in Sri Lanka. PhD thesis, University of Nottingham.

[thumbnail of Thesis-Eranga.pdf] PDF (Thesis - as examined) - Repository staff only - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
Download (4MB)

Abstract

Proso millet (Panicum miliaceum L.) is a minor cereal cultivated as a subsistence crop in Africa and Asia. It is an essential crop in rainfed farming systems of dry areas in Low Country Sri Lanka where rainfall is the limiting factor for crop production. The crop’s shorter lifespan, low water requirement and considerable yield in unfavourable environments are important adaptive features. It shows promise for high nutrient content and yield under marginal environmental conditions and is suggested as a crop that might be less affected by climate change. Proso millet in Sri Lanka is cultivated from local accessions/landraces as there are no varieties bred for specific traits. There is ample evidence to suggest that Sri Lanka's climate has already changed, and its agricultural sector is vulnerable to climate shocks. Therefore, it is important that these minor millets should receive research attention, which is the reason for selection of the crop. This study was designed to analyse the current status of Proso millet cultivation in Low Country Dry Zone Sri Lanka and to model the sensitivity of yield to climate change in mid-21st century through crop-climate modelling.

The present situation of Proso millet cultivation in Sri Lanka was initially studied using a combination of past data, on-farm field survey, field observations and field sampling approaches. Biomass samples were collected at harvesting stage from farmers’ fields. High spatial variability was observed among fields due to different land preparation techniques, quality of seeds, seed rates and plant densities. The yields reported by farmers (ranged from 0.470 to 1.956 t ha–1) were found to be far below the potential yield expected by the Department of Agriculture Sri Lanka (4 t ha–1).

Growing season rainfall characteristics and their relationship with crop husbandry strategies in growing area were analysed to study the link between Proso millet cultivation and agro-climatic characteristics in the growing area in Sri Lanka. The growing season (15th March–15th June) rainfall characteristics, onset, retreat and length of the season, and trends of dry spells in the growing area were analysed using daily rainfall data collected from 5 meteorological stations for the period of 1983–2015. It was found that Proso millet cropping patterns in the study area coincided with the distribution of rainfall and that farmers adjust crop management practices and crop selection to minimise climate associated damages to cultivation.

Agricultural Production Systems Simulator (APSIM), which has been tested under different growing conditions, genotypes and environments across the world, was selected to simulate Proso millet production in Sri Lanka. Five farmer selections (hereafter mentioned as accessions) named as L_1, L_11, L_12, L_14 and L_25 and that had been collected from farmers’ fields in Low Country Dry Zone Sri Lanka were tested. The data gathered from two field experiments (2016 July–October: dry season and 2017 February–April: rainy season) conducted at Sabaragamuwa University (6.70°N, 80.79°E) were used for model calibration. Observed crop phenology, leaf area index, biomass and yield from calibration experiments correlated well with the simulated values in the calibrated model suggesting a good calibration. The model was validated using the yield data collected from farmers’ fields grown under different sowing dates, plant densities, soil properties and crop management practices. Simulation of grain yield was satisfactory (Root Mean Square Error 279 kg ha–1) when compared with farmers’ data, therefore, the model was used to study the climate sensitivity and future yield projections of Proso millet in Sri Lanka.

The future (2040–2069) climate of the Proso millet growing region in Sri Lanka was studied using data from 20 general circulation models of Coupled Model Intercomparison Project (CMIP5) under two greenhouse gas emission scenarios (RCP4.5 and 8.5). These data suggest a clear increase in temperature and a possibility of a wetter future in the mid-21st century in the area. The sensitivity of the crop to changes in temperature and precipitation was studied systematically using the protocol of the Coordinated Climate-Crop Modelling Project (C3MP of AgMIP: The Agricultural Model Intercomparison and Improvement Project). Simulated yields tend to increase with increases in rainfall if temperature increases by less than 2 ºC, but all other combinations of change result in decreases in yield. Future climate mean yields simulated by APSIM were lower than the simulated yields using the baseline (1980–2009) climate under both RCP4.5 (–7.3%) and RCP8.5 (–16.6%) scenarios for 2040–2069 period though these changes were not significantly (p > 0.05) different from the baseline yields. Since the yields of major crops like rice are expected to have lower yields under similar climate change scenarios, Proso millet, that is grown with low inputs, without irrigation and fertiliser (most fields), will be an alternative crop that has the potential to be cultivated in Low Country Dry Zone Sri Lanka for mid-21st century under a changing climate.

This is the first attempt to evaluate the climate sensitivity of Proso millet using crop-climate modelling approaches, therefore, this study can be used as a baseline study to assess the climate sensitivity of other underutilised crops. The results can be used in decision making regarding the expansion of Proso millet cultivation in the future.

Item Type: Thesis (University of Nottingham only) (PhD)
Supervisors: Ashfold, Matthew
Karunaratne, Asha
Walker, Sue
Keywords: crop modelling, climate change
Subjects: Q Science > QH Natural history. Biology
Faculties/Schools: University of Nottingham, Malaysia > Faculty of Science and Engineering — Science > School of Biosciences
Item ID: 56888
Depositing User: Wimalasiri, Ganwari Eranga Madushan
Date Deposited: 23 Jul 2019 03:29
Last Modified: 07 May 2020 11:16
URI: https://eprints.nottingham.ac.uk/id/eprint/56888

Actions (Archive Staff Only)

Edit View Edit View