African Association of
Remote Sensing of the Environment
 
 

Articles

  • 12 Oct 2012 10:00 AM | Administrator (Administrator)

    Lien entre enneigement et climat à grande échelle sur le Haut Atlas Marocain à l’aide
    des images MODIS (Période 2000-2010) 
    A.MARCHANE, L JARLAN, L.HANICH, et A.BOUDHAR

    Faculté des Sciences et Techniques Guéliz, Département des Sciences de la Terre, laboratoire de Géoressources-Unité associée au CNRST (URAC42), BP 549, 40 000 Marrakech, Maroc. 
    E-mail : marchane.gat@gmail.com
    Centre d’Etudes Spatiales de la Biosphère, 18, Av. E. Belin 31401 Toulouse, France 
    Faculté des Sciences et Techniques, BP 523, Béni-Mellal, Maroc

    Résumé :
    Dans les pays du sud de la Méditerranée et du Moyen Orient, l’eau est une ressource rare: 180 millions d’habitants disposent de moins de 1000 m d’eau par an et par habitant et 80 millions sont en situation de pénurie (moins de 500m/habitant /an) alors que la demande a doublé depuis 50 ans. Pour de nombreux bassins versants en méditerranée semi aride, la zone de production de l’eau est située en montagne et la présence de neige constitue un réservoir non négligeable. Malgré l’importance de la neige dans le bilan hydrique régional et comme indicateur de changement climatique, peu d’études se sont focalisées jusqu’à présent sur l’étude du lien entre l’enneigement et le climat à grande échelle.

    De longues séries temporelles de données satellitaires existent désormais. Dans les régions semi-arides qui ne disposent pas d’un panel opérationnel de suivi au sol (réseaux de mesures météorologiques …), ces séries d’observation représentent un témoin unique de l’évolution des ressources qui n’a été que peu exploité jusqu’à présent en Méditerranée. Dans ce contexte, nous avons caractérisé la variabilité spatio-temporelle du couvert neigeux sur le Haut Atlas à partir des produits journaliers de couverture neigeuse acquises par l’instrument MODIS (figure 1). Plusieurs traitements basés sur des filtrages spatio-temporel (J. Tong et al. 2009 ; J. Parajka et al, 2010) visant à éliminer les pixels nuageux en particulier sont appliqués et évalués. Enfin, certaines caractéristiques saisonnières (durée d’enneigement, date des premières neiges et surface enneigé maximum) sont extraites.

    Figure 1: Cycles annuels d’enneigement (moyenne de 15 jours) sur le Haut Atlas du Tensift pour la période 2000-2010 (produits MODIS MOD10A2) 

    Dans un 2ème temps, nous avons étudié le lien entre, d’une part, la dynamique atmosphérique et océanique à grande échelle vue à travers l’oscillation Nord Atlantique (NAO) et les températures de surface de l’océan atlantique (SST) et, d’autre part, la variabilité du couvert nival sur cette région.

    Les résultats montrent que la variabilité interannuelle est extrêmement forte en terme de cycle annuel d’enneigement, de surface enneigée maximum et de date des premières neiges. L’étude du lien entre la dynamique atmosphérique et océanique à grande échelle (NAO et SST) et la variabilité du couvert enneigé sur cette région, montrent notamment qu’une NAO en phase négative durant le mois de février est légèrement favorable à l’enneigement sur le Haut Atlas. La faiblesse de cette relation,
    même si son signe négatif est cohérent avec les études précédentes montrant le lien NAOprécipitations sur l’Afrique du Nord est étonnant. L’analyse de la relation entre la NAO et la température mesurée à la station d’Oukaimeden à 3200m d’altitude a permis de montrer que la phase négative de la NAO conduisait dans un même temps à un réchauffement, donc plutôt défavorable à l’enneigement. Ce double impact opposé de la NAO sur la température et les précipitations, deux facteurs régissant le déclenchement d’un évènement neigeux, pourrait expliquer la faible relation obtenu entre NAO et enneigement. Enfin, nous avons montré une relation forte entre les anomalies de température de surface des océans dans les régions tropicales et équatoriales à la fin de l’été et l’enneigement sur le Haut Atlas. Ces résultats corroborent notamment de précédentes études et font des SST dans cette région de l’Atlantique un candidat potentiel intéressant pour la prévision précoce du couvert neigeux sur le Haut Atlas.

    De manière générale, cette étude montre l’intérêt de l’utilisation des données de la 
    télédétection pour caractériser l’enneigement en montagne semi aride et permet de déterminer les facteurs synoptiques qui régissent la dynamique de cet enneigement 

    Mots clés : hydrologie, télédétection, climat, semi-aride, neige, Haut Atlas

    Références : 
    - J. Tong, S.J.Déry, et P.L.Jackson, 2009, Interrelationships between MODIS/Terra remotely
    sensed snow cover and the hydrometeorology of the Quesnel River Basin, British Columbia,
    Canada, (Hydrol. Earth Syst. Sci., 13, 1439–1452, 2009) 
    - J.Parajka, M. Pepe, A. Rampini,S. Rossi,G. Bloschol, 2010, A regional snow-line method for
    estimating snow cover from MODIS during cloud cover,(Journal of hydrology, 381,203212,2010)

    Cet article a été écrit par Ahmed Marchane (università © Cadi Ayyad). Contactez-le à marchane.gat@gmail.com pour plus d'informations.

  • 11 Oct 2012 11:00 AM | Administrator (Administrator)

    Abstract

    The Gorongosa National Park, located in Central Mozambique, has been undergoing rehabilitation, after years of civil strife lasting from 1975 to 2004, which destroyed most of the national parks' physical and human infrastructure. Mount Gorongosa was proclaimed part of Gorongosa National Park by a government decree published in May 2010 and is the main source of water for the park. One of the biggest conservation challenges for the Gorongosa National Park, is that Mount Gorongosa, geographically detached from the core park, located in north west of the original 4,000 square kilometre park, and one of the few remaining rainforests of eastern Africa, is under threat from excessive deforestation, due mainly from increasing human settlement. The objective of the research was therefore to evaluate the land use change process on and around Gorongosa Mountain from 2000 to 2010, so as to determine the rate, state, magnitude and possible impacts of forestry loss on the Gorongosa Mountain. Land use change was analysed using Landsat satellite images of the years 2000, 2005, 2010. An IKONOS image of 2010 was used for ground truthing and this was complemented with available field survey data. Change detection was carried out on the images for the different year intervals. From the classified images, it was revealed that Montane forest declined by 13% from 29% in 2005 to 16% of the total mountain area in 2010 and this was largely attributed to cutting down of trees for cultivation purposes. Wooded grasslands, however increased from 3% to 22% and this was attributable to the practice of slash and burn agriculture around the mountain as most of the wooded grassland areas were areas of vegetation regeneration after the areas had been abandoned. For the years 2000 and 2005 an analysis of net vegetation changes particularly focusing on the Montane forest revealed that more area was lost to wooded areas and cleared areas in the magnitude of 1.5% and 3.8% respectively. Losses to cultivated areas contributed marginally in the region of 0.3%. However, between 2005 and 2010, cultivated areas, wooded grassland, and cleared areas contributed 5.5%, 4% and 1 % respectively to the loss in Montane forest on the mountain. The most dominant farming activity threatening the mountain is the cultivation of potatoes on the rich fertile mountain slopes. Overall, the continuous loss of protective vegetative cover from the mountain will not only results in loss of flora and fauna, but will also result in drying of springs, soil erosion, flush floods, loss of animal habitats, reduced water flow and siltation of Lake Urema. This will eventually leads to loss of livelihoods as people will not be able to access the traditional non timber forest products from the mountain. There is therefore an urgent need to preserve the remaining forest areas on the mountain so as to protect the water sources of Lake Urema, hence protect the “pulse” of Gorongosa National Park. There is also need to intensify reforestation activities especially on the highly sensitive areas like river sources especially to combat erosion. Since there are people already residing within the park, there is need to intensify conservation education measures so as to protect the biodiversity on the mountain and on the plains.


    ---

    This post was written by Sam Kusangaya (Gorongosa National Park). Contact him at kusangayas@yahoo.com for more information.

  • 11 Oct 2012 7:00 AM | Administrator (Administrator)

    Observation par les mages satellites des impacts du transfert de la gestion forestière aux communautés de base : cas de la commune de Didy, région d’Alaotra-Mangoro Madagascar

    RAKOTONIAINA S., RAKOTOMANDRINDRA P., RANAIVOARIMANANA S., RAKOTONDRAOMPIANA S.

    Laboratoire de Géophysique de l’Environnement et Télédétection, Institut et Observatoire de Géophysique d’Antananarivo (IOGA), Université d’Antananarivo, BP 3843, Antananarivo-101, Madagascar

    Dans le cadre d’un projet de gestion durable des ressources naturelles pour la conservation des régions hotspot de la biodiversité à Madagascar, le projet COGESFOR (Conservation et Gestion des Ecosystèmes Forestiers Malgaches) et l’Institut et Observatoire de Géophysique d’Antananarivo (IOGA) ont collaboré ensemble. Les objectifs principaux du projet sont de produire d’une part des cartes multidates de l’occupation du sol et d’analyser d’autre part l’évolution temporelle de cette dernière (détection des changements) au cours des dates prises comme des références afin d’évaluer l’efficacité du programme de transfert de la gestion forestière aux communautés de base. Le site d’étude est la commune de Didy, dans la région d’Alaotra-Mangoro, centre Nord de Madagascar. Les images satellitaires utilisées ont été celles acquises par le satellite Landsat TM 5 de résolution spatiale de 30 mètres à trois dates (1994, 2001 et 2009). Les descentes sur terrain ont été effectuées au cours de l’année 2011 avec pour objectif de relever les points GPS relatifs aux vérités-terrain correspondant aux différentes catégories d’occupation du sol existantes et de mener une enquête auprès de la population locale sur l’historique de l’environnement forestier dans leur zone. Les différentes catégories d’occupation du sol retenues sont les suivantes : forêt, savoka, tavy, zetra, rizière, savane, lac/rivière et sol nu. En plus de ces différentes classes, on a considéré aussi les classes nuages et ombre dans le cas où ces dernières sont présentes dans l’image. Les vérités-terrain acquises ont été divisées en 2 groupes : le premier en tant que parcelles d’apprentissage pour entraîner l’algorithme de classification et le second en tant que parcelles de contrôle pour valider les résultats de classification. Les images acquises à différentes dates ont été tout d’abord soumises à des corrections atmosphériques. Les classifications d’images qui ont été effectuées à partir des six bandes originales (à l’exception de la bande thermique) de chaque image multispectrale de Landsat ont été ensuite réalisées à l’aide de la méthode de classification non paramétrique SVM (Séparateur à Vaste Marge). Enfin, le module LCM (Land Cover Modeler) du logiciel de traitement d’images Idrisi a été appliqué à nos images classifiées pour détecter les différents changements apparus sur les occupations du sol entre les différentes dates considérées. Une approche d’aménagement concerté global a été définie, testée et mise en œuvre dans le site à travers de ce qu’on appelle organes de Transfert de Gestion (TG). Pour évaluer les actions entreprises au sein des TG au cours des trois années d’étude, trois principaux indicateurs écologiques ont été examinés : les surfaces forestées, les surfaces cultivées et les surfaces brulées. La transition des classes vers la classe ‘forêt’ indique ainsi un apport positif des organes de transfert de gestion (TG) sur la conservation des zones forestières et celle vers la classe ‘tavy’ (traduisant les surfaces brûlées) une évaluation de la dimension de l’exploitation des terrains par la population locale (défrichement). Les résultats statistiques montrent en général une nette diminution des surfaces forestières avant l’installation des TG dans la commune de Didy (2001) et une forte augmentation de ces surfaces après installation des TG (soit entre 2001 et 2009). Par ailleurs, la surface des zones classées ‘tavy’ a diminué au cours de ces années d’étude dans presque toute la zone d’étude.


    ---

    This post was written by Solofoarisoa Rakotonianina (Université d’Antananarivo). Contact him at solofoarisoa@gmail.com for more information.



    Full text document

  • 10 Oct 2012 3:00 PM | Administrator (Administrator)
    KEY WORDS: Urban management system (UMS), space technology, remote sensing, geographical information system (GIS).

      ABSTRACT

      Urban evolution is a situation that concerns a very significant problem in today’s world. There is a rapid growth in urban areas especially in coastal regions due to many factors depending on the specificity of each country.  Such situation often times have both socio-economic and environmental implications.   Therefore, studies in the field of urbanization which incorporates socio-economic and environmental factors for good town planning using space technology, geographical information system (GIS) and cartography in cities are of utmost importance.  Studies of this nature are lacking in most African countries and are necessary for the improvement of the quality of life. Dakar the capital of Senegal and a major city in the West-African coastal region was used as a case study.  It has a population of 2,275,351 inhabitants representing 51 % of the national population.  Studies have revealed that Dakar has the highest urban change in Senegal due to better basic infrastructure compared to the hinterland. The aim of the paper is to demonstrate how GIS techniques, cartography and high resolution remote sensing data can be incorporated into a decision support tool for utilization in urban management system (UMS).  The urban management issues considered in solving some of the present problems posed by rapid urbanization include proper allocation of habitation zone and well defined data infrastructure.  The analysis further explains the expected results and suggests some re-orientation in the application of urban management system.


      ---

      This post was written by Diallo Ngagne (Cadastre). Contact him at diallo.ngagne@gmail.com for more information.

    1. 10 Oct 2012 1:00 PM | Administrator (Administrator)

      To be presented at the AARSE 2012:

      Manuela Hirschmugl, Roland Perko, Claudia Hörmann, Ursula Schmitt, Mathias Schardt

      1. Joanneum Research, Austria, manuela.hirschmugl@joanneum.at  
      2. Technical University of Graz, Austria

      KEY WORDS: Forest monitoring, Forest mapping, REDD, SAR and optical data integration

      ABSTRACT

      This study presents an innovative processing chain and various test results for using both SAR (PALSAR) and optical (such as AVNIR, LANDSAT) data in an integrative way for forest mapping in the Congo Basin. Three activities are described in this context: (i) geometric adjustment of SAR and optical data by automatic image matching; (ii) analysis of various pre-processing steps for SAR data and (iii) a method for efficiently classify SAR data based on an existing optical classification. In part one, the robust and fully automatic matching procedure based on the mutual information maximization principle has proven to be useful to ensure geometrical congruence between optical and SAR data sets. Results show that the RMSE is reduced from over 80 m to less than 10 m without manual interaction. The second analysis covers the wide variety of SAR pre-processing methods and options. From over hundred different options, the best processing steps are selected by using cross-correlation analysis and in addition considering the typical MMU needed for forest monitoring in the tropics. The third development concerns the so-called classification-based trainer. This method allows filling classification gaps caused by clouds or sensor failures in optical data by using SAR data without much manual effort. A first benchmarking test involving AVNIR and PALSAR shows a slight overestimation of 0.8 % of non-forest area for the resulting classification compared to the classification on optical data only. Considering the difference in data quality and properties between optical and SAR data, this is a very promising result.


      ---

      This post was written by Roland Perko (Joanneum Research), Contact him at roland.perko@joanneum.at for more information.


      Full text document

    2. 09 Oct 2012 1:41 PM | Administrator (Administrator)

      A plentiful source of revenue within the aquaculture industry of Africa, mussels and oysters are ideal platforms as it concerns water quality testing.  Along the coast of South Africa the European blue mussel (mytilus galloprovinciallis) is highly prevalent.  This bivalve, which is susceptible contaminants, is ideal to serve as a testing platform to begin a Mussel Watch program within the continent of Africa.

      The Mussel Watch Program is one of the longest running contaminant monitoring programs in the coastal ocean with more than 20 years of data. Mussel Watch uses bivalves (Mussels, Oysters, and Zebra Mussels) as a means to assess water quality. The purpose of the program is geared towards assessing contaminants nationally.  Utilizing formats such as Geographic Information Systems (GIS) and Remote Sensing data assessment, this project identifies possible releasers of effluent waste into the major coastal watershed regions pertaining to ongoing research conducted within monitored Mussel Watch sites.  This project further serves as a platform for testing the mussel, a major source of income for some Coastal African nations

      The categorization of possible contaminating locations is made available through the development of a large dataset. This dataset utilizes those derived from agencies such as the United States Environmental Protection Agency (USEPA) and other federal government databases such as the National Oceanic and Atmospheric Administration (NOAA), and the United States Geological Survey (USGS).  Utilizing platforms such as ESRI® ArcMap™ software, spatially referenced locations, via point data, vector data, line data, and polygons depicting points and sites of interest were created using latitude and longitude information. Points and areas of interest (AOI) were verified using remote sensing imagery. As such, Polybrominated Diphenyl Ethers (PBDEs) within observable mussel watch sites were assessed by NOAA’s Center for Coastal Monitoring and Assessment (CCMA). Using this data, researchers are able to identify possible sources of contributors to the present contaminant. 

      In the attempt to identify possible contributors of the PBDEs contaminant with NOAA’s National Status & Trends Mussel Watch Program a suite of software was utilized in assessing and compiling the acquired data. ArcGIS 9.3 was used as the primary software in manipulating the dataset used in this project. Remote Sensing imagery acquired from an ESRI™ database served as the base map within the project and was used to verify Points of Interest (POI).

      Data collected from the USGS, was created and edited to show coastal watersheds reflective of Mussel Watch Program and observed coastal regions.  A national dataset was collected from the USEPA.  Wastewater treatment facilities were closely examined as contaminant release sites due to their potential to release untreated wastewater. Other contributors of contaminated water sources that were identified within this study included brownfields, superfund sites, power plants, hazardous waste sites, unidentified that comprised information regarding active National Pollutant Discharge Elimination System (NPDES) permits permitted facilities, and combined sewer overflows.  The data gathered was analyzed and checked for irregularities, corrected, and projected using World Geodetic System (WGS 1984).


      ---

      This post was written by Patrina L. Bly (Elizabeth City State University). Contact her at patrina_bly@yahoo.com for more information.

    3. 09 Oct 2012 7:54 AM | Administrator (Administrator)
      First reported at a nursery in Mpumalanga province in 1990, Fusarium circinatum is a fungal pathogen causing widespread mortality of Pinus radiata and Pinus patula seedlings. Improved methodologies for early disease detection are thus pertinent, and rely on identifying specific wavebands that correspond to specific physiological responses of the plant to stress. The objectives of this research were to i) determine the earliest possible window period, from time of infection, for disease detection, and ii) identify the specific hyperspectral wavebands that could be used for discriminating healthy and infected seedlings. To achieve these objectives, we setup an experiment with a sample of 3-month old P. radiata seedlings (n = 150) divided into three classes; healthy (n = 50), damaged (n = 50), and infected (n = 50). Reflectance measurements for all three classes were collected using an Analytical Spectral Devices Full Range spectroradiometer, weekly over a five week period. Reflectance measurements were later analysed using a random forest with a feature selection algorithm. Results of the analysis indicate that the best possible discrimination occurs at week three (KHAT = 0.81; out of bag (OOB) error = 12.67%). The results further indicate that wavelengths in the red-edge and near-infrared regions show the most promise in discriminating the healthy, damaged, and infected classes. These results could be explained by reduction in needle chlorophyll content expressed by a shift in the red-edge toward shorter wavelengths. Furthermore, lowered near-infrared reflectance has been associated with disease-induced stress. Overall, this study provides a basis for the early detection and discrimination of infected P. radiata seedlings that could be operationalized within a nursery environment.

      ---
      Post was written by Nitesh Keshavelal Poon (Stellenbosch University). Contact him at poona@sun.ac.za for more information.

    4. 09 Oct 2012 7:00 AM | Administrator (Administrator)
      We describe a method for improving Earth observation satellite image resolution, for specific areas of interest where the sensor design resolution is insufficient. Our method may be used for satellites with yaw-steering capability, such as NigeriaSat-2. We show that, according to the slanted edge modulation transfer function (MTF) plots, the effective resolution obtained by simulated yaw-steering of a satellite yielded a 138% improvement in resolution. This result equates to obtaining a 2.1 m resolution image from a sensor designed to acquire 5 m resolution images.

      ---
      This post was written by Riaan van den Dool. Contact him at riaanvddool@gmail.com for more information.

    5. 08 Oct 2012 6:30 PM | Administrator (Administrator)

      ABSTRACT

       Land suitability analysis is a prerequisite to achieving optimum utilization of the available land resources for sustainable agricultural production. Comprehensive, reliable and timely information on agricultural resources is very necessary for a country like Kenya, where agriculture is the mainstay of our national economy. Within Kenya, the demand for rice continues to grow as more Kenyans make changes in their eating habits, and as urban population increases but the production is very low. Lack of knowledge on best combination of factors that suit production of rice has contributed to the low production. The aim of this study was to develop a suitability map for rice crop based on physical and climatic factors of production using a Multi-Criteria Evaluation (MCE) & GIS approach. The study was carried out in Kirinyaga, Embu and Mberee counties of Central and Eastern province in Kenya. Biophysical variables of soil (soil pH, soil texture and soil drainage), climate (humidity and temperature) and topography were considered for suitability analysis. All data were stored in Arc GIS 9.3 environment and the factor maps were generated. For Multi-Criteria Evaluation (MCE), Pairwise Comparison Matrix was applied and the suitable areas for rice crop were generated and graduated. The current land use / land cover map of the area was developed from a scanned survey map of the rice growing areas in the region. According to the present land use/cover map, the rice cultivated area was 13,369 ha. Finally, we overlaid the land use/cover map with the suitability map for rice production to identify differences and similarities between the present and potential land use. However, the crop-land evaluation results of the present study identified that in the study area, 75 percent of total rice crop currently being used was under highly suitable areas and 25 percent was under moderately suitable areas. The results showed that the potential area for rice growing is 86,364 ha and out of this only 12% is under rice cultivation. This research provided information at local level that could be used by farmers to select cropping patterns and suitability.


      ---

      This post was written by Joseph Kihoro Mwangi, John Njoroge, and Hunja Murage

      (Jomo Kenyatta University of Agriculture and Technology). Contact Joseph Kihoro Mwangi at kihoromike@gmail.com for more information.

    6. 08 Oct 2012 6:00 PM | Administrator (Administrator)

      Climate change became a reality. Their impacts are rampant in many parts of the world: hurricanes, persistent droughts and rising sea levels hit many people throughout the world.

      In order to investigate these disturbances, many indicators of climate change have been developed by the National Observatory on the Effects of Global Warming French (ONERC), the Environmental Protection Agency (EPA), the Organization for Economic Cooperation and Development (OECD) ... and presented by the Intergovernmental Panel on Climate Change (IPCC) and the World Meteorological Organization (WMO). An indicator represents the state of certain environmental conditions over a given area and a specified period of time. Indicators increase our understanding of the causes and effects of climate change. Environmental indicators are a key tool for evaluating existing and future programs and providing sound science for decision-making.

      Morocco is not an exception to climate change; observations have shown that all regions of the Kingdom will be affected one way or another by these changes, which will increase their vulnerability and affect the two sectors country's most important as water and agriculture. Among these regions, we chose one of Marrakesh Tensift Al Haouz because of her geographical position and limited water resources.

      So as to betray climate change in this region, a list of these indicators has been established. Between these indicators, we picked up to work on increasing the average temperature of air by the calculation of four climate indices. To do this, we relied on the maximum and minimum temperatures of the two stations study Marrakech and Essaouira using the outputs of Statistical DownScaling Model (SDSM). Subsequently, we integrated the projections of future climate of the region, given by the same model, in a Geographic Information System (GIS).

      The results of this study show an upward trend in temperatures combined with reduced rainfall, these developments are likely to increase pressure on water resources and consequently will affect agriculture and food security in the region.


      ---

      This post was written by Niama Boukachaba (Cadi Ayyad University- Marrakesh-Morocco). Contact her at niama.boukachaba@edu.uca.ma for more information.

      This site is managed by Space 4 Development