Informations et ressources scientifiques
sur le développement des zones arides et semi-arides

Accueil du site → Master → Pays Bas → 2015 → Occurences and impacts of invasive alien plant species in the Agulhas Plain and Lacustrine Wetlands, Western Cape, South Africa

International Institute for Geo-Information Science and Earth Observation (ITC) 2015

Occurences and impacts of invasive alien plant species in the Agulhas Plain and Lacustrine Wetlands, Western Cape, South Africa

Apedo, G.A. (Gloria Afua)

Titre : Occurences and impacts of invasive alien plant species in the Agulhas Plain and Lacustrine Wetlands, Western Cape, South Africa

Auteur : Apedo, G.A. (Gloria Afua)

Etablissement de soutenance : University of Twente International Institute for Geo-Information Science and Earth Observation (ITC)

Grade : Master of Science in Geo-Information Science and Earth Observation 2015

One important land cover that is being threatened globally are wetlands. The Agulhas coastal plain is located at the Southernmost point (Cape Agulhas) of Africa in the Western Cape province of SA. Due to its biodiversity , it has wetlands considered to be of international importance by the Ramsar convention. There are views that the wetlands are being damaged due to either climate change, land use and land cover ( LULC ) activities and changes and due to an increase of invasive alien plants ( IAPs ) . The IAPs which were introduced in the late sixties in the area for a wide range of uses (e.g. sand dune fixation ), are perceived to have taken over the natural “Fynbos” vegetation which is an important indigenous and unique biome in South Africa. Alien species (e.g. certain Acacia spp.) are also seen as competitors for water resources, and are blamed also for the drying up of the wetlands because of their high evapotranspiration rates. It was postulated that the increase in agricultural activities and the use of fertilizers which runoff into the waterbodies could likewise be a basis for which the invasive plants are thriving or a possible climate change and variations could be a reason , because alien species are also rather drought resistant. The Support Vector Machine supervised multispectral image classification method (or SVM) was used and tested here to classify the images from different seasons and years in order to perform post classification for change detection analysis. Field and laboratory analysis of water samples were carried out to determine the concentrations of the mineral and chemical constituents in the water courses and bodies. The results indicated a rather unique Na-Cl hydrochemical signature and water type. This is due to the proximity of the coastal area to the Ocean and the geology in the area. Linear regression applied to a > 100- year rainfall data series showed a positive slope trend which was concluded as significant since the t
- stat of 2.12 was greater than t- critical of 1.98 at 5% level of significance. However, the small positive linear trend in the temperature data records did not test significant ; with a t- stat ( 0.36) < t- critical (2.02) . The pattern of the climate (P , T ) was found to be a cyclic at multi- decadal time scales. So, persistent years with drier or warmer conditions, seem to be exchanged for sequences of wetter (a/o colder) years. The closest agreement with climate change predictions by other researchers was the reduction in precipitation during the dry season and increase during the wet season. It was observed from the land cover classification that the areal extent of the waterbodies increased during the wet season. Also a decreasing trend in cultivated land areas was measured from the image classification. This could be explained by the acquisition program of private farmlands by SANParks of the SA Government . The natural vegetation of the area especially the vegetation class “ shrubs ” is increasing and this could be attributed to the Working for Water programme. Locations with the occurrences of the alien vegetation were found both upstream (waters with very low EC values ) and downstream on the plain in waters with high salinity content, and no relation of aliens spatial distribution with water chemistry was found. In our opinion, the pollen of the IAPs are propagated predominantly by wind and settle (also due to wash off an d runoff) in the riparian zones and grow in the drainage network . Regression analysis did not indicate any relation between the IAPs and rainfall even though there was a positive relation between rainfall variation and cultivated land areas . These seasonal trends and agricultural land are however obvious. The SVM classification method on the 30m Landsat images did not enable us to detect and delineate the IAPs and monitor their spreading . Higher resolution images and more information on the spectral reflect ances of the IAPs are recommended for effective detection and monitoring of their spread.

Mots clés : Cape Agulhas plain, wetlands, South Africa, invasive alien plants, Remote sensing, land use- land cover change, Support Vector Machine (SVM), climate change

Version intégrale (ITC)

Page publiée le 2 avril 2018