Keywords --- Assessment, population dynamics, climate variability, food security drivers , statistical learning, Chad
Introduction
There is a long tradition of scholarship on the nexus between population growth and food production dating back to the early work of Thomas Malthus in the late 1700s on the relationship between population growth and resources. There are at least two ways to approach the problem of interrelationships between population growth and food production. The first perspective, called the Malthus’ approach, determines how changes in agricultural conditions affect the demographic situation. In this method, population is a dependent variable. Malthus’ main reasoning was that while human population increased exponentially, food production increased more slowly, in a linear fashion. In other words, population growth can and will outstrip the food supply (Hodgson 2004; Malthus 1803). The second perspective, called Boserup’s approach, examines the effects of population on agriculture. In this case, population is an independent variable, which, in turn, is a major factor determining agricultural development. This implies that population influences both the development of agricultural technology, which, later, shapes the productive capacities of resources (Boserup 2017; Marquette 1997).
Based on the above stated paradigms, several studies addressing the relationships between population and agricultural production have been conducted and reported in the literature. Starting in the 1960s, Boserup demonstrated that agricultural intensity increases with population pressure. Examining how people, markets, and governments are responding to rising land pressures, Jayne, Chamberlin, and Heady (2014) showed that rising rural population densities in parts of Africa are profoundly affecting farming systems and the region’s economy. Using a two-year panel dataset, Meza-Pale and Yunez-Naude (2015) studied the impact of rainfall on agricultural production and net income among rural households in Mexico. They found a consistently negative effect of rainfall on maize production, especially for rain-fed farms and small farmers. They did not, however, find clear evidence of a significant impact of rainfall on household net income. Ndamani and Watanabe (2015) analyzed rainfall variability and its relationship with crop production using data collected between 1992 and 2012 in the Lawra district of the upper west region of Ghana. Their results revealed a negative correlation between rainfall and crop production for all crops studied. Barrios, Bertinelli, and Strobl (2010), suggested that recent trends in desertification may have affected the extent of rainfall in the semi-arid areas because a reduction in vegetative cover can translate into the absence of inter-annual soil water storage and, hence, negatively impact agricultural productivity. Gbetibouo and Hassan (2005) regressed net farm revenue with climate, soil, and socio-economic data for South African field crops. Their results indicated that the production of field crops is sensitive to marginal changes in temperature when compared to changes in precipitation. Temperature increase was positively correlated with net revenue, which declined with a reduction in rainfall.
According to Henley (2010), Chad is one of the world’s top 20 producers of sorghum, and about 19% of daily nutrition in the country is derived from sorghum. The per capita consumption of dietary staples for Chad in 2007 was 389 Kcal, 103 Kcal, and 60 Kcal for sorghum, maize, and rice, respectively. According to the FAOSTAT (2018), Chad produced 991,045 tons, 443,779 tons, and 257,701 tons of sorghum, maize, and paddy rice, respectively, in 2016. Clearly, these crops are crucially important and knowledge about production drivers would greatly improve planning and monitoring of agricultural policies intended to mitigate food insecurity and improve quality of Chadians’ lives.
Like many other sub-Sahara African countries, Chad is subject to the impact of highly variable climate conditions. Its ecosystems are affected by climate change and pressure from overcrowded human populations and livestock, often resulting in significant degradation of environmental resources. Moreover, the distribution of rainfall is not uniform across the country because Chad is located in a transitional zone between a tropical climate in the south and a desert climate in the north. Deeper knowledge and more in-depth understanding of the factors affecting food production are indispensable in order to better allocate the necessary resources to mitigate or overcome food insecurity issues. This is the motivation for this study, which investigates the relationships between human (population) and environmental (precipitation) factors and agricultural production (sorghum, maize, and rice). The study sought to quantify (1) the degree of the relationship between population or precipitation and yields of the above-mentioned crops, (2) the individual effects of population and precipitation on the crops, and (3) the combined effect of population and precipitation. Specifically, we used regression analysis to determine the extent to which changes in population and precipitation explained the variation in annual production of sorghum, maize, and rice for the period 1980 through 2011.
1.1. Description of the study area
Chad is located in north-central Africa, from about 7–23º N and 13–24 º E, straddling the sub-tropical climate band called the Sahel. Its neighbors include Libya to the north, Niger and Nigeria to the west, Sudan to the east, the Central African Republic to the south, and Cameroon to the southwest. The northern part of the country extends into the arid Sahara Desert, whereas the south has a wet tropical climate. Chad embraces a great range of climates from south to north, which are influenced by the shifting of the Intertropical Convergence Zone (ITCZ). Northern Chad receives very little rainfall throughout the year. Central Chad, which is sub-tropical, has a shorter wet season between June and September (with 50 to 150 mm of rainfall per month). Southern Chad, a tropical savanna, has a wet season between April and October (with 150 to 300 mm of rainfall per month). The country receives almost no rain during the dry season months between November and March (McSweeney et al. 2006).
Synoptic, climatological, agrometeorological, and rainfall stations are the measuring systems that constitute the weather observation network in Chad. Direction des Ressources en Eau et de la Météorologie (DREM) provides daily, decadal, and seasonal forecasts intended for preparing communities for extreme weather events. Artificial rains have been used to correct pockets of drought, which have had a positive impact on agro-pastoral production in Chad according to the National Framework for Climate Services (NFCS 2013). According to the country summary from the Population Reference Bureau (PRB 2013), the population of Chad in 2013 was estimated to be 12,825,000. FAOSTAT (2013) projected that Chad’s population will increase to 33,516,000 by 2050 whereas PRB (2016) estimated it to be 38.5 million by 2050. The number of births per 1,000 people is 47, deaths per 1,000 people are 14, and the rate of natural increase is 3.3%. The infant mortality rate is 87, and the total fertility rate is 6.4 (PRB 2016).
The World Bank (2017) noted that the agriculture sector in Chad employs more than 87% of the active workforce. Funk et al. (2012) reported that the main crops are millet and sorghum in the northern part of the agricultural zone, with increased crop diversification around Lake Chad and areas further south. They concluded that human and animal pressures on a degraded ecosystem combined with limited agricultural development have led to low levels of national food production. Figure 1 represents the study area and shows the weather station distribution where the population is greater than 18,000 inhabitants. The precipitation map overlaid on the station locations was generated using precipitation data from the Tropical Rainfall Measuring Mission (TRMM 2011), the weather station data was collected from Weather in the World (2015), and population data obtained from the Mongabay website (2018).