Environmental Monitoring Namibia

Background
Our department has been cooperating with the Namibia University of Science and Technology (NUST) since middle of 2019. A first joint excursion to Namibia was conducted in September/October 2019 with master students from our M.Sc. program ‘Resource Analysis and Resource Management’ in conjunction with NUST students from the Faculty of Natural Resources and Spatial Sciences. A Memorandum of Understanding (MoU) between GAU and NUST was signed in August 2020 providing a base for further cooperation and applied research between both institutions.

Namibia is a land of contrast from a socio-economic as well as environmental perspective and is located on the southwestern coast of Africa. Namibia is twice the size of Germany but has a total population of only 2,6 Mio (2021) resulting in one of the world’s lowest population densities. Apart from breathtaking landscapes and landforms, Namibia faces a series of challenges such as poverty due to inequality in income distribution throughout the country, proper management of natural resources to sustain agriculture, mining and fisheries, the COVID situation which has strongly impacted the tourism sector as a main source of income as well as ongoing social segregation and spatial marginalization in the post-apartheid regime. Other challenges are the growth of informal settlements around larger urbanizations through increasing population and rural-urban migration and a slow land reform process. A major challenge are the effects of climate change requiring sustainable adaptation and mitigation strategies. Namibia is a semi-arid country with long dry periods and short episodes of abundant and highly erratic rainfall during the rainy season (October until April) caused through a Shift of the ITC zone or through the periodic El Niño- und La Niña – phenomena. Extreme weather conditions such as floods or extensive droughts have increased in the past years due to climate change with strong impact on surface and ground water availability, rangeland and agricultural productivity, food security and further land degradation such as bush encroachment or soil erosion. These conditions especially impact livelihoods in the northern communal areas as most people live in subsistence economy which is closely connected to hydrological conditions. The past 10 years were characterized through a perennial drought lasting from 2013 to 2016 and an extreme drought occasion during the rainy season of 2018/2019, which was the driest in 90 years. In contrast, January 2021 saw rainfall totals double to triple the norm. Due to climate change, an increase of exceptional droughts due to late and poor rainfall performance is expected.

MODIS Color Composites of parts of the Cuvelei-Etosha-Basin (CEB) in Northern Namibia showing the Etosha salt pan in January 2019 (left) and January 2021 after excessive rainfall (right). A significant greening of vegetation can be seen in the right image where abundant rainfall has led to an increased ephemeral conversion of the dry salt pans into wet landscapes covered by a thin layer of water. (Source: NASA WorldView Snapshots)

Current research:


Current research activities are wrapped around three areas of interest dealing with drought monitoring, grassland monitoring, hydrological modelling and health.

    The expected increase of exceptional droughts due to late and poor rainfall performance will strongly affect agricultural productivity, grazing capacity and water availability in Namibia with severe impacts on rural health and household food security. Drought monitoring is an important task to support suitable response measures by supervising changing dynamics of soil, water and vegetation cover. The early detection of droughts potentially reduces negative impacts on water resources and the agricultural sector. This can be achieved through satellite imagery and derived products in various resolutions allowing detailed as well as supra-regional time series analysis and monitoring. There are many ways to map and monitor drought based on more than 100 existing drought indicators. This makes it extremely difficult to select the right indicator for a specific situation or region. Research has been undertaken to compare various drought indices based on medium resolution MODIS (Moderate Resolution Imaging Spectroradiometer) satellite imagery using terrestrial vegetation greenness, evapotranspiration and surface temperature products. The indices provide possibilities of time series analysis and cross-comparison to identify, visualize, monitor and better understand the nature, characteristics and spatial-temporal patterns of drought in northern and central Namibia.

    Study area and land cover based on ESA Copernicus Global Land Service in a 100m spatial resolution
    Study area and land cover based on ESA Copernicus Global Land Service in a 100m spatial resolution. The Land Cover Classification shows mainly shrubland with open forests in the north as well as wetlands (Zambezi, Okavango delta) and herbaceous vegetation surrounding the Etosha and Makadikgadi salt pans.
    Source: Wyss et.al., 2021


    The below example shows two selected MODIS-based indices: VCI (Vegetation condition index) and TCI (Temperature condition index) for the past five rainy seasons (September to April) based on an underlying 20-year time series (2001 to 2021). VCI represents the current vegetation activity compared to the historical range. Low index values indicate stressed vegetation and high values represent above normal conditions, which stands for healthy vegetation conditions. TCI can be used to determine temperature related vegetation stress and estimate the soil moisture content. The illustration shows that droughts happen every year but have different levels of classified drought intensity. Indices show similar spatial patterns but different levels of classified drought intensity with a longitudinal increase of index values from East to West and a latitudinal increase from North to South following the rainfall gradient and therefore highly correlate with ERA-5 precipitation reanalysis data provided through the Copernicus Climate Change Service. The extreme drought year of 2018/2019 is clearly visible.

    representation of seasonal VCI (Left), TCI (Middle) and ERA-5 precipitation reanalysis data showing mean seasonal precipitation in mm.
    Side by side representation of seasonal VCI (Left), TCI (Middle) and ERA-5 precipitation reanalysis data showing mean seasonal precipitation in mm. (Right) for a 5-year period 2016/17 until 2020/21.
    Source: Wyss et.al., 2021


    Future research will focus on more detailed inter-annual analysis of droughts within identified drought-prone areas (hotspots) following a multi-sensor approach incorporating higher resolution satellite imagery such as Sentinel 2/3 and Landsat 8/9 to derive further biophysical properties. The possible integration of data on soil characteristics, soil moisture in various spatial resolutions as well as socio-economic investigations could support the establishment of a more integrated drought index to assist decision makers in regional drought assessments and mitigation efforts.

    Mean deviation of ESA SMOS (Soil Moisture and Ocean Salinity)
    Mean deviation of ESA SMOS (Soil Moisture and Ocean Salinity) data over a 10-year period (2010 to 2020) showing dryer areas with negative values such as the area around Oshakati (Cuvelai Etosha Basin) in the upper north-west and wetter areas with positive values (i.e. Okavango Delta). Source: Rettner, 2021.

    Grassland Monitoring:
    Since 2023, our department has engaged in a joint research project with NUST to monitor grasslands using a multi-sensor approach. Project area is ProNamib, an area of interesting ecological transition between the Namib Desert and Nubib Mountains covering a broad spectrum of habitats consisting of sand and gravel plains, stretches of woodland savanna, mountain ranges and vegetated dune belts. The NamibRand Nature Reserve and Pro-Namib Nature reserves (PNNR) form part of the greater Sossusvlei-Namibrand landscape, an MEFT (Ministry of Environment and Tourism) driven initiative to foster conservation management across a large part of the central-southern Namib Desert. The climate is semi-arid with an annual mean rainfall of 100-150mm, but up to 200mm in the Pro-Namib Nature Reserve. ProNamib aims at further land acquisition in order to facilitate seasonal migratory wildlife routes and to protect biodiversity. Grassland restoration, through free movement of animals into areas which have received higher levels of precipitation, is a major aspect of biodiversity conservation in this area. Especially as biomass has declined during the last eight years of perennial and extreme drought events during the growing seasons. It is expected that re-establishment of historic grazing will lead to a significant increase of biomass and thus create a significant new carbon sink in the coming years. Research will effectively link remote sensing technologies to existing ecosystem conservation strategies for grassland restoration and support sustainable grassland management strategies through creation of a status quo vegetation map and further monitoring of biophysical properties and forage quality of grassland within the research areas under changing environmental conditions

    Link to the website of the ProNamib Nature Reserve

    Remote sensing (hyperspectral, multispectral) technology has proven to be extremely useful for monitoring spatial-temporal patterns of grassland health with new satellite generations holding great promise to provide more accurate and timely information on the greenness (activity) of vegetation. It is envisaged that the project links-up with ecosystem conservation strategies, such as best land management practices for grassland restoration, to support adaptation to increasing droughts brought on by climate change. A central activity and objective in the project are the measurement of field spectra using an institutional ASD-FieldSpec3 spectroradiometer in order to create a unique spectral library for the most important grassland species and types. The measurements will support the development of a suitable approach to combine field spectroscopy and multispectral/hyperspectral satellite imagery through data fusion techniques for mapping, monitoring, and estimation of relevant proxies for grassland condition, composition and biomass. So far, no hyperspectral library for grass types has been developed in Namibia. It is therefore important to establish basic information on the spectral behavior of grasses throughout their phenological phase. The hyperspectral benefits will allow for a better (i) detection of mixtures of materials within same pixel, (ii) identification of specific materials with high degree of accuracy, (iii) get some measure of relative abundance based on depth of absorption features, (iv) producing quantitative (rather than qualitative) result about grass type mixture. The targeted integration and data fusion of new emerging hyperspectral satellite generations (EnMAP, PRISMA) with multispectral imagery (Sentinel-2, Landsat) and field spectroscopy holds great promise to support the PNNR grassland rehabilitation efforts through provision of more accurate and timely information on phenological status and grassland development in the NamibRand and Pro-Namib nature reserves. Carbon sequestrations are currently being carried out by two NUST postgraduate students on honors and master level in 2021 and 2022 for a total of 92 random sample plots with a spatial extent of 25x25 m, stratified according to topographic conditions (riverbed, grass plains, mountains), of which 82 samples lie within the extent of the acquired PRISMA hyperspectral scenes (see figure 1). The upscaling and pixel-based resampling to match band definitions and resolutions of our hyperspectral and multispectral sensors will allow the evaluation of statistical relationships between vegetation biomass and field spectra. In addition, information gathered through NUST (carbon sequestration) will effectively be used for calibration, statistical cross validation with derived image products and algorithm trainings. As a side-product, multi-temporal timeseries analysis will be performed based on suitable multispectral and hyperspectral vegetation indices as well as biophysical properties (LAI, FPAR) using multispectral (Sentinel, Landsat) and hyperspectral (EnMAP, PRISMA) satellite imagery. Results will provide a better understanding of the intra- and interannual spatial distribution, key biophysical properties, phenological status and development of grasslands in this semi-arid region over time and will be correlated to Reanalysis precipitation data such as (ERA-5, CHIRPS) as a driving factor for vegetation growth and vitality. Field spectra and gathered in-situ data will again be used to evaluate derived vegetation indices for estimation of grassland yield and quality.

    : Location of the Pro-Namib nature reserve (PNNR) with the Namib desert to the West and Nubib Mountains to the East containing NUST sample plots for carbon sequestration.
    Figure 1: Location of the Pro-Namib nature reserve (PNNR) with the Namib desert to the West and Nubib Mountains to the East containing NUST sample plots for carbon sequestration.


    Goals and major objectives:
    The three major objectives are:
    1. Collect and document spectral profiles of dominant grass species (spectral library).
    2. Develop a suitable approach to combine field spectroscopy and multispectral/hyperspectral satellite imagery through data fusion techniques to map, monitor, and estimate relevant proxies for grassland condition.
    3. Establish correlation between spatiotemporal pattern of grasslands (quality and quantity) and patterns in environmental drivers.

    First results were published in Frontiers of Remote Sensing (see publications) und and were presented during the 43 EARSel-Syposium in Manchester (2024)

    Poster_EARSel_2

    We currently have two doctoral candidates from NUST registered at our university and working within our department. One doctoral research project research project is to identify climate change impacts by monitoring catchments behavior through satellite-based systems, global reanalysis precipitation and temperature datasets and ground observation measurements for the Okavango river shared between Angola, Namibia and Botswana; and parts of the Omuramba-Omatako rivers stretching from north-eastern to central Namibia. The project has the following main goals:

    i) To monitor and characterize catchment behavior to determine drought events using various environmental and climate indices and models.

    ii) To assess and evaluate socio-economic impacts of drought occurrences in and around catchment areas. Influences of droughts on economic and social activities (i.e. farming, tourism, industries, drinking water, etc.) are not well understood and require in-depth assessment to incorporate indigenous knowledge to existing scientific environmental and climate models. A first paper examines runoff estimation in the Okavango Omatako river basin in Namibia based on a SWAT tool (Soil & Water Assessment) to simulate water balance and runoff estimation in the Okavango Omatako river basin based on a semi-distributed hydrological model.

    A second doctoral thesis is looking at health aspects related to climate change and ongoing droughts which highly impact household food security and rural health, especially in areas depending on subsistence agriculture. The thesis will concentrate on the mapping of malnutrition prevalence in children under five years in the northern Kavango, and central Hardap regions of Namibia. Research objectives include:

    1. Identification of contextual factors that affect the nutrition status of children under five years;

    2. Analysis of contextual factors that affect the nutrition status of children under five years;

    3. Exploring links between malnutrition and contextual factors;

    4. Identification of spatial distribution pattern of malnutrition prevalence;

    5. Performing mapping and modelling of malnutrition prevalence using Geographic Information Systems (GIS).



    Publications
    1. Malena Andernach, Daniel Wyss, Martin Kappas (2020) - An Evaluation of the Land Cover Classification Product Sentinel 2 Prototype Land Cover 20 m Map of Africa 2016 for Namibia. In: Namibian Journal of Environment 4 A: 1-12.

    2. Kaleb Gizaw Negussie, Martin Kappas, Daniel Wyss, Nichola Knox, Eva Corral-Pazos-de-Provens, Miguel Vallejo Orti (2022)
    Evaluating SWAT Model for Runoff Estimation in the Semi Arid Catchment of the Okavango - Omatako River Basin, Namibia. In: African Journal of Environmental Science and Technology, 16(11)

    3. Wyss, D.; Negussie, K.; Staacke, A., Karnagel, A.; Engelhardt, M.; & Kappas, M. (2022)
    A comparative analysis of MODIS-derived drought indices for Northern and Central Namibia. In: African Journal of Environmental Science and Technology, 16(5), 173-191.

    4. Bantelmann, P.; Wyss, D.; Pius, ET.; and Kappas, M. (2024)
    Spectral imaging of grass species in arid ecosystems of Namibia. In: Frontiers of Remote Sensing 5:136855

    Contact:
    Prof. Dr. Martin Kappas
    Dr. Daniel Wyss