Lecturer: Remote Sensing/GIS
Name: Mahlatse Kganyago
Location: D2 Lab 344C Auckland Park Kingsway Campus
Geography Environmental Management Energy Studies Staff Staff Members
Tel: +27 (0) 11 559 3645
About Dr Mahlatse Kganyago
Mr Mahlatse Kganyago holds MSc degree (cum laude) in applied remote sensing and GIS and is a licenced Remotely Piloted Aerial Systems (RPAS or UAV) operator. He has extensive experience in machine learning applications, image analysis (multispectral and hyperspectral), spectral analysis (NNIR-MIR range), land-cover/use mapping, retrieval of crop biophysical and biochemical variables, statistical and spatial modelling of landscapes, and both research and teaching of Remote Sensing/Earth observation/Geographical Information Systems for natural and agro-ecological resources management. His work in recent times focuses on enhancing the use, understanding, development, and implementation of novel remote-sensing technologies and robust machine learning algorithms to aid agro-ecological systems research and application development for societal benefit. He participated in internationally funded projects such as CropWatch for South Africa funded by UK Space Agency, EU H2020 Enhancing Food Security in African AgriCultural Systems with the support of Remote Sensing (AfriCultuReS, http://www.africultures.eu/), and Wetlands Monitoring and Assessment Service for Transboundary Basins in Southern Africa (WeMAST, http://wemast.sasscal.org/) funded GMES and Africa programme (EU-AU). He is a member of the AARSE algorithms committee and previously participated in international working groups such as Committee on Earth Observation Satellites-Working Group on Capacity Building (CEOS-WGCapD) and Group on Earth Observations (GEO) Land Degradation and Neutrality (LDN). He currently works as a Lecturer (Remote Sensing/GIS) at the University of Johannesburg (UJ) in the Department of Geography & Environmental Management & Energy Studies.
- Applications of state-of-the-art machine learning algorithms for classification of vegetation and retrieval of biophysical and biochemical parameters
- Time series analysis of seasonal and long-term vegetation dynamics and climate-related extreme conditions such as wildfires, heat waves and droughts
- Multi-sensor and multi-source data fusion for enhanced spatial and temporal properties for environmental monitoring based on new data architectures such as the Data Cube
- Field spectroscopy and hyperspectral data analysis
- Interpretable machine learning algorithms and model-agnostic methods
Land degradation assessment with multi-sensor data and machine learning algorithms
Land degradation is the temporary or permanent decline in the productive capacity of the land, and the diminution of the productive potential, including its major land uses (e.g., rain-fed arable, irrigation, forests, rangelands), its farming systems (e.g., smallholder subsistence), and its value as an economic resource. Land degradation is the single most pressing current global problem, causing a reduction in productivity, loss of ecosystem services, and threatens the survival and development of humankind for present and future populations.
It is caused by a complex of interacting environmental and anthropogenic factors, such as competing socio-economic activities, poor land management practices, and climate change. The global goal is to achieve Land Degradation Neutrality (LDN), which is defined as ‘a state whereby the amount and quality of land resources, necessary to support ecosystem functions and services and enhance food security, remains stable or increases within specified temporal and spatial scales and ecosystems.’
The aims of the project are to:
- Assess the status and impacts of land degradation induced by various environmental and socio-economic factors using satellite data from various sensors and machine learning algorithms at various geographical scales; and
- Develop cutting-edge methodologies relevant for characterising various aspects of land degradation, towards LDN and support reporting for Sustainable Development Goals (SGDs) indicator 15.3.1 – Proportion of land that is degraded over total land area.
Characterisation of vegetation species, vegetation conditions, vegetation changes and essential climate variables (ECV) in various environments with new generation sensors, analysis-ready-data and field spectroscopy
The aim of this project is to characterise, through classification and regression,
- the vegetation species: invasive alien species, medicinal plants, threatened species, crop types
- vegetation conditions: pest and diseases, nutrients, drought
- vegetation changes: wildfires, clearance, poaching, bush encroachment, tree cover density
- and essential climate variables (ECVs): (above-ground-biomass, FaPAR, Canopy Chlorophyll Content, Leaf Area Index, Burned area) in various environments with new generation sensors, analysis-ready-data, and field spectroscopy.
Calibration and validation of satellite data and products to determine and improve consistency and accuracy in various environments
Globally, improving satellite characterization of geophysical and biophysical properties is of paramount importance for a variety of applications. Satellite products such as Land cover, FaPAR, LAI, above-ground biomass, and fire disturbance, are regarded as essential climate variable (ECV) by the Global Climate Observing System (GCOS). Moreover, these are critical to support the implementation and monitoring and reporting for global and regional mandates such as the United Nations Sustainable Development Goals (UN-SDGs) and African Union-Agenda 2063 (AU-A2063). The consistency and accuracy of these products are uncertain, especially in Africa, where such studies are rare. Furthermore, quantifying uncertainties in satellite products is useful for users and developers interested in the operational use of the product and further development.
The aim of this project is to assess the accuracy and consistency of various satellite data products including analysis-ready-data (ARD) and Vegetation indices and ECVs.
PUBLICATIONS IN SCIENTIFIC JOURNALS
- M. Kganyago, C. Adjorlolo, M. Sibanda, P. Mhangara, G. Laneve, T. & Alexandridis (2022) Testing Sentinel-2 Spectral Configurations for Estimating Relevant Crop Biophysical and Biochemical Parameters for Precision Agriculture Using Tree-based and Kernel-based Algorithms. Geocarto International.
- M. Kganyago, C. Adjorlolo, & P. Mhangara (2022) Exploring Transferable Techniques to Retrieve Crop Biophysical and Biochemical Variables Using Sentinel-2 Data, Remote Sensing.
- M. Gasela, M. Kganyago & G. De Jager (2022) Testing the utility of the resampled nSight-2 spectral configurations in discriminating wetland plant species using Random Forest classifier, Geocarto International.
- L. Shikwambana, M. Kganyago (2022) Analysis of wildfires and associated emissions during the recent strong ENSO phases in Southern Africa using multi-source remotely-derived products. Geocarto International.
- Shikwambana L and Kganyago M (2022) Meteorological Influence of Mineral Dust Distribution Over South-Western Africa Deserts Using Reanalysis and Satellite Data. Environ. Sci. 10:856438. doi: 10.3389/fenvs.2022.856438
- R. Pritchard, T. Alexandridis, M. Amponsah,… M. Kganyago, … Developing capacity for impactful use of Earth Observation data: Lessons from the AfriCultuReS project. Environmental Development.
- M. Kganyago, P. Mhangara, C. Adjorlolo (2021) “Estimating Crop Biophysical Parameters Using Machine Learning Algorithms and Sentinel-2 Imagery” Remote Sensing
- L. Shikwambana, M. Kganyago (2021) Assessing the Responses of Aviation-Related SO2 and NO2 Emissions to COVID-19 Lockdown Regulations in South Africa Remote Sensing
- M. Kganyago (2021) Using Sentinel-2 Observations to Assess the Consequences of the COVID-19 Lockdown on Winter Cropping. Remote Sensing Letters
- L. Shikwambana, Kganyago, M. (2021) Temporal Analysis of Changes in Anthropogenic Emissions and Urban Heat Islands during COVID-19 Restrictions in Gauteng Province, South Africa. Aerosol and Air Quality Research
- M. Kganyago, K. Govender, V. Sivakumar, L. Shikwambana (2021) Blazing Wildfires at the Outeniqua Pass in South Africa during the October/November 2018 Period. Remote Sensing Applications: Society and Environment
- M. Kganyago, L. Shikwambana (2021) Did COVID-19 Lockdown Restrictions have an Impact on Biomass Burning Emissions in Sub-Saharan Africa? Aerosol and Air Quality Research
- L. Shikwambana; M. Kganyago; P. Mhangara (2021) Temporal analysis of changes in Anthropogenic Emissions and Urban Heat Islands during COVID-19 Restrictions in Gauteng Province, South Africa.” Aerosol and Air Quality Research
- L. Shikwambana; M. Kganyago (2021) Observation of aerosol and gaseous emissions from wildfires using Sentinel-5p, CALIPSO and MERRA-2 datasets: A study of USA, Brazil, and Australia wildfires in 2018/19 Atmosphere
- M. Kganyago, G. Ovakoglou, N. Mashiyi, T. Alexandris, J. Odindi, C. Adjorlolo (2020) Validation and Inter-comparison of Sentinel-2 Leaf Area Index (LAI) Product Derived from SNAP Toolbox in an African Semi-Arid Agricultural Landscape. Remote Sensing Letters
- M. Kganyago, L. Shikwambana (2020) Assessment of the Characteristics of Recent Major Wildfires in the USA, Australia and Brazil in 2018 – 2019 Using Multi-Source Satellite Products Remote Sensing
- L. Shikwambana, M. Kganyago (2020) Trends of atmospheric pollutants from oil refinery processes: A case study over the United Arab Emirates. Remote Sensing Letters
- M. Kganyago, L. Shikwambana (2019) Assessing Spatio-temporal Variability of Wildfires and their Impact on Sub-Saharan Ecosystems and Air Quality using Multisource Remotely Sensed Data. Sustainability.
- M. Kganyago, P. Mhangara (2019) The Role of African Emerging Space Agencies in Earth Observation Capacity Building for Facilitating the Implementation and Monitoring of the African Development Agenda: The Case of African Earth Observation Program. ISPRS International Journal of Geo-Information
- OE Malahlela, C Adjorlolo, JM Olwoch, ML Kganyago, MJ Mashalane (2019) Integrating geostatistics and remote sensing for mapping the spatial distribution of cattle hoofprints in relation to malaria vector control. International Journal of Remote Sensing
- Kganyago, M., Odindi, J., Adjorlolo, C. & Mangara, P. (2018) Evaluating the Capability of Landsat 8 OLI and SPOT 6 for Mapping Invasive Alien Species in an African Savanna Landscape. International Journal of Applied Earth Observation & Geoinformation
- Kganyago, M., Odindi, J., Adjorlolo, C. & Mangara, P. (2017) Selecting a subset of spectral bands for mapping invasive alien plants: a case of discriminating Parthenium hysterophorus using field spectroscopy data. International Journal of Remote Sensing
- R. Pritchard, T. Alexandridis, M. Amponsah, N. B. Khatra, D. Brockington, T. Chiconela, J. O. Castillo, I. Garba, M. Gomez-Gimenez, M. Haile, C. Kagoyire, M. Kganyago, … (2022) Developing capacity for impactful use of Earth Observation data: Lessons from the AfriCultuReS project. Environmental Development.
- T. Alexandridis, G. Ovakoglou, I. Cherif, M. Giménez, G. Laneve, D. Kasampalis, D. Moshou, S. Kartsios, M. Karypidou, E. Katragkou, S. Garcia, M. Kganyago, N. Mashiyi, K. Pattnayak, A. Challinor, R. Pritchard, D. Brockington, C. Kagoyire, J. Beltran (2021) Designing AfriCultuReS services to support food security in Africa. Transactions in GIS
- M. Mohan, G. Richardson, G. Gopan, M. Aghai, S. Bajaj, GA Galgamuwa, M. Vastaranta, P. Pitumpe Arachchige, L. Amorós, A. Corte, S. de-Miguel, R. Vieira Leite, M. Kganyago, E. Broadbent, W. Doaemo, M. Shorab, A. Cardil (2021) UAV-supported forest regeneration: Current trends, challenges and implications. Remote Sensing
- Sibandze, P., Mangara, P., Odindi, J. & Kganyago, M. (2014) A comparison of Normalised Difference Snow Index (NDSI) and Normalised Difference Principal Component Snow Index (NDPCSI) techniques in distinguishing snow from related cover types. South African Journal of Geomatics.
LOCAL AND INTERNATIONAL CONFERENCE PUBLICATIONS/PRESENTATIONS
- M. Kganyago; A. Ramoelo; E. Zoungrana; N. Mashiyi; & I. Garba (2022) Characterizing the Spatial Distribution of Grazing and Browsing Resources in Africa Using Random Forest Classifier and Multi-Sensor Data. IGARSS 2022 – 2022 IEEE International Geoscience and Remote Sensing Symposium. 17-22 July 2022. Kuala Lumpur, Malaysia
- M. Kganyago, M. Mukhawana, M. Mashalane, A. Mgabisa, S. Moloele (2021) “Recent Trends of Drought Using Remotely Sensed and In-Situ Indices: Towards an Integrated Drought Monitoring System for South Africa” 2021 IEEE International Geoscience and Remote Sensing Symposium IGARSS
- I. Cherif, G. Ovakoglou, T. Alexandridis, M. Kganyago, N. Mashiyi “Improving water bodies detection from Sentinel-1 in South Africa using drainage and terrain data” Remote Sensing for Agriculture, Ecosystems, and Hydrology XXIII, SPIE
- I. Otte, N. Mashiyi, P. Kluter, S. Hill, A. Hirner, J. Eberle, M. Urban, A. Mlisa, M. Kganyago, M. Schwinger, U. Gessner, C. Schmullius, J. Baade (2021) “Development of earth observation data cubes for monitoring land degradation processes in South Africa” EGU General Assembly Conference Abstracts
- C. Schmullius, M. Urban, A. Hirner, C. Berger, K. Schellenberg, A. Ramoelo, I. Smit, T. Strydom, G. Chirima, T. Morgenthal, B. Melly, U. Gessner, N. Mashiyi, A. Mlisa, M. Kganyago, J. Baade (2020) “Earth Observation Strategies For Degradation Monitoring In South Africa With The Sentinels-Results From The Spaces II Saldi-Project” IGARSS 2020-2020 IEEE International Geoscience and Remote Sensing Symposium, 26. Sep. – 2. Okt. 2020, Waikoloa, Hawaii.
- M. Kganyago, G. Ovakoglou, P. Mhangara, T. Alexandridis, J. Odindi, C. Adjorlolo, N. Mashiyi (2020) “Validation of atmospheric correction approaches for Sentinel-2 under partly-cloudy conditions in an African agricultural landscape” Remote Sensing of Clouds and the Atmosphere XXV, SPIE
- M. Kganyago “AfriCultuReS Livestock Service: An operational effort towards monitoring the status and productivity of grasslands in arid and semi-arid regions of sub-Saharan Africa” (2019) Annual Grassland Society of Southern Africa (GSSA) Conference; 30 June – 04 July 2019, 1 Olifantshoek Road, Keidebees, Upington, Northern Cape, South Africa.