生態系と生態学のジャーナル

抽象的な

Likoti Farming under Changing Climate in Lesotho: Agronomic Grain Yield versus Technical Efficiency

Olaleye AO, Tambi E, Bangali S and Odularu GOA

Climate Change (CC) and with sub-optimal nutrient contents in the soils of Lesotho is negatively impacting yield and yield components of maize. Often, the grain yield is often ≤ 2.50 t/ha in most cases. One of the ways smallholders in Lesotho try to mitigate the impact of CC and sub-optimal nutrient contents is through the practise of Conservation Agriculture (CA) called likoti farming. Data (soil and socio-economic variables) were collected from smallholder farmers practising Likoti Farming Systems (LFS) in Lesotho. Surface soil (0-20 cm) samples and socioeconomic data were collected between 2012/2013 and 2013/2014 cropping seasons from 105 smallholder farmers in 22 villages. These villages are located in four administrative districts of Lesotho (Berea, Butha Buthe, Leribe and Maseru). The socio-economic variables were on inputs (farm size, payment for land, quantity of fertilizer, fertilizer cost, quantity of seeds, man-days, wage rate, quantity of herbicides, price of herbicides and labour cost) and output (grain yield) used by the farmers. In addition, long-term monthly rainfall data (1900-2007) were collected from the Lesotho Metrological Services. Socio-economic data were analysed using means procedure of SAS across the villages. In addition, these same data (i.e., inputs and output) were subjected to Data Envelopment Analysis (DEA) using SAITECH DEA-Solver under constant returns to scale (CRS). Results showed that soils across these villages/ districts had sub-optimal contents of N, P, K and acidic pH (i.e., ≤ 5.0). The mean annual rainfall has been declining steadily over the years. Examination of the grain yields across villages showed that highest grain yield (e.g. 2113.0 kg/ha) was recorded in Ha-Ts’alemoleka (Butha Buthe District) and the least was about 100 kg/ha in Ha-Khoeli village (Maseru district). When the DEA method was used to examine the same data set using inputs and output, only four (or 18.20%) of the LFS/Decision Making Units (DMUs) were technically efficient (i.e., efficiency score was 1.0), while the other 18 DMUs (or 81.80%) were not efficient. It was observed that agronomic research using grain yields (i.e., output) as the reason why soil conservation is better in one village/region compared to others may be erroneous. Hence, efforts should be made by researchers (i.e., agronomist, soil scientists, animal scientists, extension agents etc..) to use DEA software to evaluate collected data (i.e., input variables) along with the output(s) arising from such trials to make proper decisions to policy makers.