Climate change in North-Western Cambodia: impact assessment, vulnerability and scope for adaptation in upland agricultural systems

Abstract

Increases in mean temperature, changes in rainfall patterns and increases in frequencies of extreme climate events have put more pressure on agriculture to meet the world’s food and fibre demand. Agricultural food produce amongst smallholder farms has a vital role in feeding the world population, particularly in developing countries. Over 80% of the food that is consumed in the developing world is produced by small-scale farmers. In general, small-scale farmers are unable to achieve potential yields due to inability to manage various yield-limiting factors. Predicted impact of climate change is one of the major threats to sustained agricultural productivity, especially for rainfed production systems. It is, therefore, critical to reduce constraints to achieving crop yield potential in order to minimise the adverse impacts of climate change. This study was conducted in North-West Cambodia where Pailin Province (12°52' N, 102°36' E) and Samlout District in Battambang Province (12°40' N, 102°45' E) were selected to examine climate change impacts and adaptation options for smallholder farms in the region. Locally relevant climate change projections for Battambang city (13°05' N, 103°13' E), 80 km from the study areas, were downscaled from Global Circulation Models (GCMs). The objectives of this study were to (1) better understand upland rainfed cropping systems and socio-economics of smallholder farms, (2) assess farmers' perceptions of climate variability and change against regional climate records and projections, (3) evaluate farmers’ autonomous adaptation options to short‐term climate change, and (4) investigate options for longer‐term planned adaptation measures. The thesis progresses through the following sequence. The initial stage of this study involved collection of baseline data from smallholder households on cropping systems, socio-economic context, production constraints, climate change perceptions, and perceived impacts and employed adaption options to climate change. The data were gathered from 390 representatives of households of the selected study areas in North-West (NW) Cambodia. The results showed upland crop production has a critical role in supporting the livelihoods and food security of the local people where about 70% of family income is obtained from crop production. These farmers have implemented crop intensification practices on small pieces of land to maximise their incomes. The farmers often face various significant challenges to their productivity and income such as: adverse climate impacts; soil fertility depletion; biotic constraints; high input costs; and unpredictable fluctuations in crop prices. For a five-year period between 2008 and 2012, farmers had experienced crop yield losses as perceived reductions of 16-27%. These yield reductions had posed a significant threat to the livelihoods, food security and welfare of the people. This has been in line with the rising number of young people leaving their home villages for casual employment in cities and outside the country. The baseline study also revealed that the local people perceived warming temperature, rising frequency and severity of drought and dry spells, declining annual rainfall, and a later start and later finish to the wet season. The variations and changes in climate parameters were perceived to have significant effects on the farmers’ crop productivity. The baseline study led to further research on climate change assessment and crop simulation modelling for the selected study areas. The climate assessment study involved gathering available climate records and downscaling climate change projections from 23 GCMs for the study locations. The records reveal that temperature has already become warmer, and is projected to continue warming over the remainder of the 21st century with average warming rates per decade of about 0.13oC and 0.24oC under intermediate emission (RCP4.5) and high emission scenarios (RCP8.5), respectively. The records for the current annual rainfall also showed no clear trend, but more fluctuations between years. Similarly, projections for past rainfall indicated neither downward nor upward trends, but exhibited a slight upward trend after the 2010s onwards. Furthermore, projections showed more rain would fall in the main wet season (May-October) and some would be in the very early dry season (November), while the rainfall was projected to remain low in the dry season (December-May). These findings should be taken into account in planning adaptation measures because farmers firstly need to perceive climate change correctly before they can employ effective actions to mitigate and adapt to climate change. Crop simulation modelling based on the downscaled climate change projections was used to examine climate change impacts on the current crop production strategies and explored various feasible adaptation measures for the study areas. Simulation modelling showed the current practice of planting the first crop in the late dry season (February/March) had failure rates of around 59% at Pailin and 32% at Samlout. In addition, the expected failure rates would be higher if soil fertility depletion was included in the simulation. The study suggested that farmers could grow two good crops of maize per year (<1% chance of crop failure at both sites) when the first crop is planted in May and the second in September. Compared with the current sowing times, the proposed sowing windows (the first crop planted in May and the second in September) showed increases in annual gross margin (US$/ha) of 35% ($415) and 689% ($560) for Pailin and Samlout, respectively. Further findings indicated that soil fertility depletion was a critical factor limiting crop profitability where crops were planted at the safer times. The study revealed two feasible options, additional fertiliser application and crop rotation, which have the potential to address the profitability losses. Compared with the current crop sequence under the improved sowing windows, the average annual gross margins per hectare under fertiliser addition for Pailin and Samlout explained increases of 137% ($1717) and 230% ($2084), respectively. Similarly, on average annual gross margins for crop rotation (with soybean, mungbean and sunflower), same comparison, increased by 140% ($1739) for Pailin and 218% ($2009) for Samlout. It is expected that not every proposed climate change adaption option will be adopted by local farmers, as they might have their own ideas managing climate variability and climate change. In addition, smallholder responses to climate change are likely to be influenced by other socio-economic challenges. Therefore, the proposed adaptation options from the simulation modelling were presented to local farmers for consideration. Farmers agreed that a number of response actions such as fertilizer addition, crop rotation, growing legumes, retaining crop residues, minimizing cultivation, adjusting sowing dates, changing crop species and using improved seed quality would reduce crop yield and financial losses. Most of these proposed measures, are not currently implemented. Farmers were reluctant to adopt new practices they have not implemented before. These farmers lack governmental extension advice and relevant information to help them take action to improve crop productivity. In other words, there is no governmental extension service at the village level, and farmers depend on each other and their value chain network, such as input suppliers and traders, for advice on crop decision-making. This study recommends that future interventions include the local value-chain network actors to improve the chance of adoption. The results from the study are also useful to policy makers and researchers as well as other stakeholders (such as international donors, NGOs and private sector actors) for project planning, implementation and assessment at the regional and sub-regional levels.