Learning from history of natural disasters in the Sahel: a comprehensive analysis and lessons for future resilience.

One of the first environmental crises to attract interest in development initiatives and aid was the great drought of the 1970s in the Sahel. This study investigates the extent of damage caused by natural disasters from one of the most widely used databases-EM-DAT-with a sample size of 16 Sahelian countries over the period 1960-2020. These countries have been divided into three regions: Western Africa Sahel (WAS), Central Africa Sahel (CAS), and Eastern Africa Sahel (EAS). The analyses encompass four categories of natural hazards, namely, biological, climatological, hydrological, and meteorological. We used descriptive and test statistics to summarize the natural disaster records. Through this approach, we explore tendencies to identify the most frequently reported natural hazards; we examine their spatial distribution and evaluate their impacts in terms of socioeconomic damage and causalities. During the study period, a total of 1000 events were recorded in the database. The Western Africa Sahel (WAS) region had the highest number of disasters, with 476 events, followed by the Eastern Africa Sahel (EAS) region with 369 events. The most common hazards in the Sahel were hydrological (41.8%), mainly floods, and biological (39.5%) hazards. Approximately 300 million people in the Sahel were affected by natural hazards, with 59.17% in EAS, 36.48% in WAS, and 4.35% in CAS. Although droughts occurred less frequently (14%), they had a significant impact on the population, affecting 84% of those affected by natural hazards. In general, EAS experiences a higher impact from natural hazards, potentially influenced by the pastoral lifestyle of its population. However, WAS is also very vulnerable to natural hazards especially epidemics and nowadays floods. The uncontrolled urbanization in the area may contribute to this vulnerability.

State and parameter estimation for a class of schistosomiasis models.

We develop a general framework to estimate the proportion of infected snails and snail-human transmission parameter of a class of models that describes the evolution of schistosomiasis. To do so, we consider simultaneously the dynamics of schistosomiasis, captured by the homogeneous version of the classical MacDonald's model, and the measurable output: the number of female schistosomes per single host. The proposed method consists of designing an auxiliary dynamical system, called observer, whose solutions converge exponentially to those of the system capturing the schistosomiasis model. Moreover, we derive an estimation of the snail-human transmission rate, an unknown but key parameter in the dynamics of schistosomiasis. These estimations are central in two of the strategies of controlling schistosomiasis, namely the use of molluscicides and mass drug administration. To further investigate control strategies on a larger scale, we consider a heterogeneous model which consists of an arbitrary number of human groups or patches and an arbitrary number of fresh-water sources, natural habitats of snails. Provided that the data of infected humans' worm burden in each patch or group is available, we provide a method of estimating the proportion of infected snails in each snail natural habitat, thereby providing a map on where to implement control strategy to mitigate or eliminate Schistosomiasis.

On the stock estimation for a harvested fish population.

We consider a stage-structured model of a harvested fish population and we are interested in the problem of estimating the unknown stock state for each class. The model used in this work to describe the dynamical evolution of the population is a discrete time system including a nonlinear recruitment relationship. To estimate the stock state, we build an observer for the considered fish model. This observer is an auxiliary dynamical system that uses the catch data over each time interval and gives a dynamical estimate of the stock state for each stage class. The observer works well even if the recruitment function in the considered model is not well known. The same problem for an age-structured model has been addressed in a previous work.

An observer for a nonlinear age-structured model of a harvested fish population.

We consider an age-structured model of a harvested population. This model is a discrete-time system that includes a nonlinear stock-recruitment relationship. Our purpose is to estimate the stock state. To achieve this goal, we built an observer, which is an auxiliary system that uses the total number of fish caught over each season and gives a dynamical estimation of the number of fish by age class. We analyse the convergence of the observer and we show that the error estimation tends to zero with exponential speed if a condition on the fishing effort is satisfied. Moreover the constructed observer (dynamical estimator) does not depend on the poorly understood stock-recruitment relationship. This study shows how some tools from nonlinear control theory can help to deal with the state estimation problem in the field of renewable resource management.