Groundwater availability is critical to the Umbeluzi Catchment. Currently, there is a need for a simple tool that can asses the availability of resources in the ground.
This especially to asses the permits for groundwater extractions. It is expected that a simplified modelling approach can provide a trend analysis sufficient for the water authorities in Mozambique to perform assessments of the sub-surface water availability. Furthermore, the water availability will be assessed for current and future conditions, under different scenarios of climate change and demand increase.
Within the project, FutureWater will develop a groundwater model in WEAP, using the Strategic Model previously build for the Umbeluzi catchment. To this end a detailed data gathering activity will take place proceed by developing the model. We aim to validate and improve the model with measurements available of groundwater levels in the catchment. The model will be validated with the technical team of ARA-Sul. Ultimately, a dedicated training session for ARA-SUl will ensure that model operation is performed by local experts.
In our ongoing commitment to bolster the efforts of ARA-Sul in Mozambique, FutureWater recently conducted an intensive training course focusing on the application of the Strategic Water Allocation Model within the Umbeluzi Catchment area. This significant initiative entailed the utilization of the renowned Water Evaluation and Planning System (WEAP) model, coupled with a comprehensive update of critical information and underlying assumptions.
The primary objective of this training was to empower the dedicated professionals at ARA-Sul with the knowledge and skills necessary to effectively manage and optimize water resources within the region. The strategic allocation of water resources is of paramount importance, especially in areas like the Umbeluzi Catchment, where water plays a pivotal role in sustaining livelihoods, ecosystems, and economic activities.
One key aspect of this training involved fine-tuning the analysis-scenarios to comprehensively assess potential bottlenecks and challenges within the water allocation system. Identifying these bottlenecks is essential for making informed decisions, developing mitigation strategies, and ensuring the sustainable utilization of water resources.
Our collaborative efforts with ARA-Sul extend beyond the training itself. We are committed to providing ongoing support and guidance to ensure the long-term success of this endeavor. Through regular follow-up activities and consultations, the technical professionals at ARA-Sul are now well-equipped to independently maintain their model and conduct the essential analyses required for informed decision-making.
Eswatini’s development is at risk by natural drought hazards. Persistent drought is exacerbating the country’s existing challenges of food security and the ability to attain sustainable development. Therefore, FutureWater, Hydrologic, and Emanti Management joined forces to bring together technologies and complementary expertise to implement the GLOW service which includes: short-term and seasonal forecasts of water availability and demand, an alerting service when forecasted water demand is higher than water availability, and water distribution advisories to reduce impact and maximise water security for all water users.
The GLOW service will be piloted in the Maputo River and Mbuluzi River Basins where three-quarters of the population of Eswatini lives, which includes the Hawane dam that supplies water to Mbabane (Capital City of Eswatini) and which is the major water supply source for Maputo, a Delta city (1 million inhabitants) which suffers from water shortages. The main beneficiaries of this project are the Joint River Basin Authority (JBRAS-PB) and the 5 River Basin authorities, AraSul (Mozambique) and the Department of Water and Sanitation (South Africa).
The innovation of GLOW is bringing together proven and award-winning technologies of advanced earth observation, open data, high-performance computing, data-driven modelling, data science, machine learning, operations research, and stakeholder interaction. These technologies require minimum ground truth information, which makes them very scalable and applicable in poorly monitored environments throughout the world. The coherent combination of the technologies into one decision support service ensures the optimum division of water, basically distributing every drop of water to meet the demands of all interests present in large river catchments.
The practice of using remote sensing imagery is becoming more widespread. However, the suitability of satellite or flying sensor imagery needs to be evaluated by location. Satellite imagery is available at different price ranges and is fixed in terms of spatial and temporal resolution.
TerraFirma, an organization in Mozambique with the task to map and document land rights, hired FutureWater, HiView and ThirdEye Limitada (Chimoio, Mozambique) to acquire flying sensor imagery over a pilot area near Quelimane, Mozambique. The objective of this pilot project is to determine the suitability of using flying sensor imagery for cadastre mapping in an area of small-scale agriculture in Mozambique.
Flying sensor imagery is adaptable and can be deployed at any requested time. The suitability of these remote sensing approaches is piloted in this study for a small-scale agricultural area in Mozambique. A pilot area is used as case study with flights made during a period of a few days in December 2020, by local flying sensor (drone) operators in Mozambique (ThirdEye Limitada).
The flying sensor imagery was acquired over the period of a few days in December 2020, for a total area of 1,120 hectares. This imagery was used as input for various algorithms that can be suitable for classification and segmentation, namely R packages (kmeans, canny edges, superpixels, contours), QGIS GRASS segmentation package, and ilastik software. This study shows some initial results of using flying sensor imagery in combination with these algorithms. In addition, comparison is made with high resolution satellite imagery (commercial and publicly available) to indicate the differences in processing and results.
With the conclusions from this pilot project, next steps can be made in using flying sensor imagery or high resolution satellite imagery for small-scale agriculture in Mozambique. The time and effort needed for the delineation of field boundaries can be largely reduced by using remote sensing imagery and algorithms for automatic classification and segmentation.
For smallholder farming systems, there is a huge potential to increase water productivity by improved (irrigated) water management, better access to inputs and agronomical knowledge and improved access to markets. An assessment of the opportunities to boost the water productivity of the various agricultural production systems in Mozambique is a fundamental precondition for informed planning and decision-making processes concerning these issues. Methodologies need to be employed that will result in an overall water productivity increase, by implementing tailored service delivery approaches, modulated into technological packages that can be easily adopted by Mozambican smallholder farmers. This will not only improve the agricultural (water) productivity and food security for the country on a macro level but will also empower and increase the livelihood of Mozambican smallholder farmers on a micro level through climate resilient production methods.
This pilot project aims at identifying, validating and implementing a full set of complementary Technological Packages (TP) in the Zambezi Valley, that can contribute to improve the overall performance of the smallholders’ farming business by increasing their productivity, that will be monitored at different scales (from field to basin). The TPs will cover a combination of improvement on water, irrigation, and agronomical management practices strengthened by improved input and market access. The goal is to design TPs that are tailored to the local context and bring the current family sector a step further in closing the currently existing yield gap. A road map will be developed to scale up the implementation of those TPs that are sustainable on the long run, and extract concrete guidance for monitoring effectiveness of interventions, supporting Dutch aid policy and national agricultural policy. The partnership consisting of Resilience BV, HUB, and FutureWater gives a broad spectrum of expertise and knowledge, giving the basis for an integrated approach in achieving improvements of water productivity.
The main role of FutureWater is monitoring water productivity in target areas using an innovative approach of Flying Sensors, a water productivity simulation model, and field observations. The flying sensors provide regular observations of the target areas, thereby giving insight in the crop conditions and stresses occurring. This information is used both for monitoring the water productivity of the selected fields and determining areas of high or low water productivity. Information on the spatial variation of water productivity can assist with the selection of technical packages to introduce and implement in the field. Flying sensors provide high resolution imagery, which is suitable for distinguishing the different fields and management practices existent in smallholder farming.
In May 2020, FutureWater launched an online portal where all flying sensor imagery from Mozambique, taken as part of the APSAN-Vale project, can be found: futurewater.eu/apsanvaleportal
Project video: Portrait of the activities on water productivity
Water and food security are at risk in many places in the world: now and most likely even more in the future, having large economic and humanitarian consequences. Risk managers and decision-makers, such as water management authorities and humanitarian-aid agencies/NGOs, can prevent harmful consequences more efficiently if information is available on-time on (1) the impact on the system, economy or society, and also (2) the probabilities for a failure in the system. EO information has proven to be extremely useful for (1). For looking into the future, considering the uncertainties, novel machine learning techniques are becoming available.
The proposed development is incorporated into an existing solution for providing Drought and Early Warning Systems (DEWS), called InfoSequia. InfoSequia is a modular and flexible toolbox for the operational assessment of drought patterns and drought severity. Currently, the InfoSequia toolbox provides a comprehensive picture of current drought status, based mainly on EO data, through its InfoSequia-MONITOR module. The proposed additional module, called InfoSequia-4CAST, is a major extension of current InfoSequia capabilities, responding to needs that have been assessed in several previous experiences.
InfoSequia-4CAST provides the user with timely, future outlooks of drought impacts on crop yield and water supply. These forecasts are provided on the seasonal scale, i.e. 3-6 months ahead. Seasonal outlooks are computed by a novel state-of-the-art Machine Learning technique. This technique has already been tested for applications related to crop production forecasting and agricultural drought risk financing. The FFTrees algorithm uses predictor datasets (in this case, a range of climate variability indices alongside other climatic and vegetative indices) to generate FFTs predicting a binary outcome – crop yields or water supply-demand balance above or below a given threshold (failure: yes/no).
The activity includes intensive collaboration with stakeholders in Spain, Colombia and Mozambique, in order to establish user requirements, inform system design, and achieve pilot implementation of the system in the second project year. Generic machine learning procedures for training the required FFTs will be developed, and configured for these pilot areas. An intuitive user interface is developed for disseminating the output information to the end users. In addition to development of the forecasting functionality, InfoSequia-MONITOR will be upgraded by integrating state-of-the art ESA satellite data and creating multi-sensor blended drought indices.
The scope of the project work is as follows:
Train selected NCBA Clusa PROMAC staff on drone operation, imagery processing software, and crop monitoring;
Provide technical assistance to trained NCBA Clusa staff on drone operation, imagery processing, and interpretation of crop monitoring data;
Present technical reports on crop development and land productivity (i.e. crop yield) at the end of the rainy and dry season
The trainings and technical assistance for the NCBA Clusa staff are provided in collaboration with project partners HiView (The Netherlands) and ThirdEye Limitada (Central Mozambique). Technical staff of the NCBA Clusa are trained in using the Flying Sensors (drones) in making flights, processing and interpreting the vegetation status camera images. This camera makes use of the Near-Infrared wavelength to detect stressed conditions in the vegetation. Maps of the vegetation status are used in the field (with an app) to determine the causes of the stressed conditions: water shortage, nutrient shortage, pests or diseases, etc. This information provides the NCBA Clusa technical staff and extension workers with relevant spatial information to assist their work in providing tailored information to local farmers.
At the end of the growing season the flying sensor images are compiled to report on the crop development. The imagery in combination with a crop growth simulation model is used to calculate the crop yield and determine the magnitude of impact the conservation agriculture interventions have in contrast with traditional agricultural practices.
Twiga’ is the Swahili word for ‘giraffe’, a keen observer of the African landscape. TWIGA aims to provide actionable geo-information on weather, water, and climate in Africa through innovative combinations of new in situ sensors and satellite-based geo-data. With the foreseen new services, TWIGA expects to reach twelve million people within the four years of the project, based on sustainable business models.
Africa needs reliable geo-information to develop its human and natural resources. Sixty percent of all uncultivated arable land lies in Africa. At the same time Africa is extremely vulnerable to climate change. Unfortunately, the in situ observation networks for weather, water, and climate have been declining since the 1970s. As a result, rainfall predictions in Africa for tomorrow have the same accuracy as predictions in Europe, ten days ahead. To realize the tremendous potential of Africa while safeguarding the population against impacts of climate change, Earth observation must be enhanced and actionable geoinformation services must be developed for policy makers, businesses, and citizens. New in situ observations need to be developed that leverage the satellite information provided through GEOSS and Copernicus (Open data/information systems).
TWIGA covers the complete value chain, from sensor observation, to GEOSS data and actionable geoinformation services for the African market. The logic followed throughout is that in situ observation, combined with satellite observations and mathematical models, will result in products consisting of maps and time series of basic variables, such as atmospheric water vapour, soil moisture, or crop stage. These products are either produced within TWIGA, or are already available with the GEOSS and Copernicus information systems. These products of basic variables are then combined and processed to derive actionable geo-information, such as flash flood warnings, sowing dates, or infra-structural maintenance scheduling.
The TWIGA consortium comprises seven research organisations, nine SMEs and two government organisations. In addition it uses a network of 500 ground weather stations in Africa, providing ready-to-use technical infrastructure.
FutureWater’s main role in TWIGA is centered around the use of flying sensors to map crop conditons, flood extent, and energy fluxes, complementing and improving data from in situ sensors and satellites. Furthermore, FutureWater is involved in innovative app development.
Nowadays, projects that invest in sustainable water management and agriculture require evidence that the targeted measures to boost water productivity are effective. Water productivity monitoring therefore becomes increasingly important. Water productivity requires data on yields and water consumption (evapotranspiration). Yield data are often difficult to obtain from farmers, especially in areas with many smallholders. Evapotranspiration is even more difficult to assess in the field. Remote sensing-based and model-based monitoring of water productivity has a large potential, also to identify yield gaps and assess the local feasible effectiveness of measures.
The objective of this pilot study was to achieve plot-level maps of water productivity and yield to test a methodology to assess the performance of different farmers in order to provide them with recommendations to improve water productivity. More specifically, this pilot study combined high-resolution imagery from Flying Sensors (FS) with a crop water productivity model to assess yield and water productivity for several plots with maize in Mozambique. Canopy cover was derived from the imagery and linked with the crop model simulations to obtain water productivity maps covering the entire growth cycle. The methodology is also used for the monitoring of crop performance during the growth season and can be used to forecast yield by the end of the season.
This feasibility study demonstrated that there is an opportunity to further develop a service that monitors water productivity based on FS-imagery and crop modelling. Service costs outweigh the additional revenues obtained by farmers. The experimental development has demonstrated that the service is technically feasible and can provide the tangible outputs needed. To bring the proposed service to a higher level of maturity, it is recommended to focus future development activities on (i) Testing for different locations and crops, (ii) Further enhancing link between FS-based imagery and crop modelling, and (iii) Involving end-users and testing within a project where WP-measures are implemented.
Background
A key factor in enabling an increase and efficiency in food production is providing farmers with relevant information. Such information is needed as farmers have limited resources (seed, water, fertilizer, pesticides, human power) and are always in doubt in which location and when they should supply these resources. Interesting is that especially smallholders, with their limited resources, are in need of this kind of information. Spatial information from flying sensors (drones) can be used for this. Flying sensors offer also the opportunity to obtain information outside the visible range and can therefore detect information hidden for the human eye (Third Eye). Nowadays, low-cost sensors in the infra-red spectrum can detect crop stress about two weeks before the human eye can see this.
The ThirdEye project supports farmers in Mozambique and Kenya by setting up a network of flying sensors operators. These operators are equipped with flying sensors and tools to analyse the obtained imagery. Our innovation is a major transformation in farmers’ decision making regarding the application of limited resources such as water, seeds, fertilizer and labor. Instead of relying on common-sense management, farmers are now able to take decisions based on facts, resulting in an increase in water productivity. The flying sensor information helps farmers to see when and where they should apply their limited resources. We are convinced that this innovation is a real game-changing comparable with the introduction of mobile phones that empowered farmers with instantaneous information regarding markets and market prices. With information from flying sensors they can also manage their inputs to maximize yields, and simultaneously reduce unnecessary waste of resources. In summary, the missing information on markets has been solved by the mobile phone introduction, the flying sensors close the missing link to agronomic information on where to do what and when.
Flying sensors
Thanks to our innovation, farmers’ demand for key agricultural information will be satisfied by means of an extension service based on flying sensor (drone) information. The deployment of flying sensors is unique in its ability to provide farmers with real-time, high-resolution, and on-demand information. We provide essential agricultural information:
At an ultra-high spatial resolution (NDVI)
With unprecedented flexibility in location and timing
Based on wavelengths not observable by the human eye
With a country-specific business oriented approach.
To this end, we use low-cost high-resolution flying sensors (drones) in a development context to ensure that farmers will get information at their specific level of understanding, and simultaneously develop a network of service providers in Mozambique and Kenya.
A flying sensor is a combination of a flying platform and camera. Reliable flying sensors are on the market in a wide-range of categories each with its specific characteristics. Based on the consortium’s experiences over the last years low-cost flying sensors have been identified that are excellent equipped for our innovation. Typically, a flying sensor flies at a height of 100 meter and overlapping images are taken about every 5 seconds. This results in individual images covering about 50 x 50 meter and an overlap of 5 images for each point on earth. So, in order to cover 100 ha 500 images are taken during a flight.
The use of Flying Sensor is unique and no comparative techniques exist that provide farmers with real-time high-resolution information. The use of satellites to provide farmers with spatial information has been promoted but has three main limitations: they have fixed overpass times, the spatial resolution is low, and the presence of clouds halters the information. It is unlikely that, within the coming decades, progress in satellites will be real competitors of flying sensors. Another category of comparable techniques to provide farmers with information is the use of ground sensors. Typical examples of these sensors are soil moisture devices, soil sampling and laboratory analysis, crop sampling and laboratory analysis. However, all those sensor techniques have the common limitation that information is only local point representative, while the main question farmers have is regarding to spatial differences. Moreover, these ground sensors are in all cases too expensive to be used by small-scale farmers.
Our flying sensors have cameras which can measure the reflection of near-infrared light, as well as visible red light. These two parameters are combined with a formula, giving the Normalized Difference Vegetation Index (NDVI). This information is delivered at a resolution of 2×2 cm in the infra-red spectrum. Infra-red is not visible to the human eye, but provides information on the status of the crop about two weeks earlier than what can be seen by the red-green-blue spectrum that is visible to the human eye. NDVI is the most important ratio vegetation index and says something about the photosynthesis activity of the vegetation. Moreover, NDVI is an indicator for the amount of leaf mass, and therefore, ultimately biomass. In general, open fields have a NDVI value of around 0.2 and healthy vegetation of around 0.8. NDVI values give an indication of crop stress. This can be caused by a lack of water, lack of fertilizer, pests or abundancy of weeds.
This Flying Sensor is equipped with infrared sensors that detect crop stress about two weeks before the naked eye can observe this.
NDVI technology
When light falls on a leaf, reflection occurs. The amount of reflection of green light (0.54 µm) is very high, making plants green to the human eye. Healthy vegetation does not reflect much red light (0.7 µm), since it is absorbed by chlorophyll abundant in leaves. In the near-infrared spectrum (0,8 µm) the amount of reflection increases rapidly to 80% of the incoming light. This increase is caused by the transition of air between cell kernels. This is characteristic for healthy vegetation.
Damaged plant material does not show this increase in reflected near-infrared light. Moreover, the reflection of red light is much higher than in healthy plant material. By measuring the reflection in these spectra, damaged plant material can be distinguished from healthy plant material (Schans et al., 2011).
With our NDVI technology damaged plant material can be distinguished from healthy plant material.
Progress
From 2014 to 2017, FutureWater has been granted support from the Securing Water for Food program, funded by USAID, Sida and the Dutch Government of Foreign Affairs, for piloting the use of flying sensors to support farmers in Mozambique with their decision making in farm and crop management. In Mozambique, we have transferred our technical skills to local ThirdEye operators over the past 3 years. We currently have 6 active local operators providing service to more than 3,500 farmers over more than 1,600 ha. These operators are able to support over 400 small-scale farmers, by collecting information and sharing it with farmers on weekly basis. Based on the information, farmers take decisions on where to do what in terms of irrigation, fertilizer application and pesticides, helping them improve their water productivity. Furthermore, they now have the capacity to deal with technical issues and are very skilled in providing advice to farmers. As a result, the farmer’s water productivity was increased by 55%, meaning less water is used to achieve the same crop yield as without ThirdEye services. ThirdEye has evolved since 2014 from a start-up to becoming the leading company in Mozambique in the field of mapping and monitoring services for farmers based on aerial images, which will continue to expand its activities over the coming years.
Since last year, the ThirdEye service is also implemented in Kenya as part of the Smart Water for Agriculture program implemented by SNV. Last month the first round of training was given to 5 operators, who will be serving at least 2,000 smallholder farmers the coming months. Training consists of flying sensor use, technical skills, safety and protocols, imagery processing and consultancy. After this, the operators will start working regularly in the regions of Meru and Nakuru. Here they will go the farmer’s fields, conduct flying sensors flights, process the images and give advice on improving their agricultural practices. Next to the service for smallholder farmers, ThirdEye delivers various services to medium and big sized farmers.
