Generating High-Resolution Digital Elevation Models for Wetland Research Using Google Earthtm Imagery – an Example from South Africa

Generating high-resolution digital elevation models for wetland research using Google EarthTM imagery – an example from South Africa

E Hoffmann and F Winde* North-West University, Potchefstroom Campus, School of Environmental Science and Development, Private Bag X6001, Potchefstroom 2520, South Africa

Abstract

Digital elevation models (DEM) generated in geographical information systems (GIS) have proven to be useful tools in hydrological research, aiding, amongst others, the delineation of catchment areas, identification of drainage patterns and flow pathways as well as for runoff determinations. They are of particular value in areas of comparatively flat topography, where such tasks are often difficult to perform. However, owing to the fact that elevation differences in wetlands typically fall below or just into the range of contour intervals of standard topographic maps being generally 20 m, and 5m for some areas, the latter fail to give enough detail. This means that sufficiently detailed relief information is often difficult to obtain for wetland research. Site- specific, high-resolution relief surveys are too expensive, relative to many research budgets, to constitute a viable alternative. Based on an approximately 12 km² study area surrounding a karst-related peatland, this paper presents a method to retrieve the required high-resolution elevation data at 1 m intervals, at low cost, from satellite imagery in Google EarthTM. The paper describes procedures used to capture and process the data using GIS ArcDesktop™ to produce a high-resolution contour map and DEM. For quality assurance purposes the generated map is visually compared to 5 m and 20 m contour intervals of standard topographic maps (1:50000) issued by the Chief Directorate Surveys and Mapping (CDSM). After correcting an off-set of 5 m it was found that the deviation of the generated contour map based on Google data from CDSM contours was in the same order as the deviation between the 2 CDSM contour sets. Finally, all 3 contour maps were compared to a contour map of 0.5 m-interval resolution specifically generated for the study area using aerial photography from an airborne survey. This too con- firmed the overall good reliability of the generated, Google EarthTM-based, contour map. Although Google EarthTM’s contour models are based on data of the Shuttle Radar Topography Mission (SRTM), the direct use of freely available SRTM data for localised, high-resolution DEM yielded unsatisfactory results. This may be due to (unspecified) procedures, or unknown data sources employed by Google EarthTM that enhance the quality of relief data. With updating intervals of 2 to 4 years, satellite imagery in Google EarthTM offers the additional advantage of containing much more recent information on relevant hydrological features than the outdated topographic maps available for the study area. It is concluded that the presented method allows the generation of high-resolution DEMs especially for areas of flat topography where adequate relief information is either not available or too costly to generate. These DEMs are useful for further wetland research.

Keywords: digital elevation model, DEM, Google Earth, GIS, wetland, hydrology, relief, contour intervals, Gerhard Minnebron peatland, SRTM, topographic maps

27°0'0"E 27°15'0"E 27°30'0"E 27°45'0"E 28°0'0"E 28°15'0"E Introduction Wonderfonteinspruit catchment Owing to potential threats, associated with continued and poorly-regulated peat min- GMB ing activities, to the ecological integrity of Randfontein Kerkskraal Dam a peatland in the semi-arid karst region of Mohlakeng

the North-West Province (South Africa), 26°15'0"S nspruit 0 10 20 Km r Wonderfontei

v the Department of Water Affairs and i Westonaria 1:500,000 R Khutsong

i Carletonville Forestry (DWAF) (now the Department of oo M Welverdiend Water Affairs (DWA)), in conjunction with Blyvooruitzicht  the Department of Agriculture (DoA) initi- Gerhard Minnebron Deelkraal ated a research project to assess impacts of 26°30'0"S peat extraction on the hydrological func- Legend

D Canals near Gerhard Minnebron tioning of the Gerhard Minnebron (GMB) Boskop Dam u Gerhard Minnebron catchment To it Sp ru Non-perennial stream it peatland (see Fig. 1). 1 0 5 R perennial stream

Dams near Gerhard Minnebron

Potchefstroom Settlements

Wetlands 26°45'0"S * To whom all correspondence should be Gerhard Minnebron Dolomite addressed. wetland  +2718 299 1582; fax: +2718 299 1582; 0 2,500 5,000 m e-mail: [email protected] Received 20 April 2009; accepted in revised form Figure 1 12 November 2009. Location of study area and upstream hydrography

Available on website http://www.wrc.org.za 53 ISSN 0378-4738 (Print) = Water SA Vol. 36 No. 1 January 2010 ISSN 1816-7950 (On-line) = Water SA Vol. 36 No. 1 January 2010

E E

° °

7 8 2 2 96 km

26°S

r e Figure 2 v i ruit R teinsp Example of a (triangular irregular i erfon o Wond o network-based) DEM for an M area of approximately 8 700

m km² surrounding the studied k 1 GMB 9 wetland at Gerhard Minnebron wetland (GMB), generated from 20 m contour intervals extracted from 16 topographical 1:50000 scale maps, comprising all of the sheets of quadrants 2627 N A through D, as issued by the Height difference: 220 m CDSM

27°S

In the course of the study a GIS (geographic information wetlands with their characteristically small changes in relief. system) was implemented to spatially represent data acquired The hydrology of the larger study area is controlled to a for this wetland. This paper focuses on how high-resolution large extent by the presence of highly karstified dolomite which elevation data acquisition and integration into a GIS can be hosts some of SA’s largest aquifers. The Gerhard Minnebron done cost-effectively by using virtual globe systems such as wetland itself owes its existence to the continuous outflow of Google Earth™ to model areas with little change in relief. dolomitic groundwater from the Gerhard Minnebron karst Hydrological modelling and GIS development were spring, yielding some 60 to 80 Mℓ/d. While falling outside the historically used in parallel but as separate avenues of spe- surface catchment of the Wonderfonteinspruit (WFS), which cialisation. This continued until the late 1980s, when efforts is the single largest tributary of the Mooi River, the Gerhard to improve the analytical abilities of GIS began to blossom Minnebron Spring is possibly hydraulically linked to the WFS and researchers began to draw these 2 fields closer together system via a network of underground karst channels present (Raper and Kelk, 1991; Sui and Maggio, 1999). This fusion in what is known as the Boskop-Turffontein compartment has resulted in the integration of hydrological techniques (Winde, 2006). The location of the GMB wetland in relation to by GIS program suites (Sui and Maggio, 1999). ESRI’s outcropping dolomite is shown in Fig. 1. ArcDesktop™ suite is one such a program suite and was At the Gerhard Minnebron wetland the maximum verti- selected to set up a GIS for this study. Its 3D Analyst exten- cal difference between the highest and lowest elevation points sion was of particular use and others, for example, ArcHydro, does not exceed 15 m. This means that in standard topographic show considerable value regarding the visualisation and maps, such as the 1:50 000 series (which is the highest resolu- modelling of hydrological processes and features. tion map series covering the whole of South Africa) displaying In spite of the challenges faced regarding the use and integra- 20 m contour intervals, this vertical difference in topography tion of GIS in hydrological studies, there have been numerous will be lost. For areas covered by (aerial) orthophotographs examples of such endeavours, as is noted by Sui and Maggio at 1:10 000 scale the Chief Directorate: Surveys and Mapping (1999) and Raper and Kelk (1991). Wetland modelling-related (CDSM) is able to provide derived contour information at integration examples can also be seen in studies regarding runoff 5 m intervals in digital format. While this greatly improves processes (Drogue et al., 2002; Ko and Cheng, 2004; Soulsby et the vertical resolution of DEMs at sub-regional scale it is still al., 2006), flood predictions (Al-Sabhan et al., 2003), water qual- insufficient for adequately capturing hydrologically-relevant ity concerns (Janauer, 1996; Bobba et al., 2000; Janssen et al., topographic detail in wetlands, where relief changes well below 2005; Jewitt et al., 2004), and water balance questions (Alemaw 5 m are often crucial for determining flow patterns. In South and Chaoka, 2003; Fraser et al., 2001; Lacroix et al., 2002; Africa professional consulting firms specialising in aerial sur- Portoghese et al., 2005), to name but a few cases. veys are often the only alternative source for generating A crucial component in modelling of hydrological proc- sufficiently detailed relief data. Depending on the location of esses such as surface runoff, catchment delineation and the the site (distance from airport) this option is commonly too determination of the nature of surface-ground water interac- costly for many research budgets. In the particular case of the tions is a DEM (digital elevation model) that can accurately 10.8 km² study area at GMB, costs for the aerial survey at capture the terrain in sufficient detail. DEMs also allow surface 0.5 m contour resolution amounted to ZAR100 000 (Minnaar, flow pathways to be identified without resorting to compli- 2009) equating to ZAR9 259/km2. cated automated hydrological modelling techniques. Relief For ’metropolitan and growth areas the CDSM provides reflected by standard topographic contours at a 20 m interval higher resolution DEMs (not contour maps) based on 50 m x 50 is acceptable for DEMs at regional scale, even for terrains with m (0.25 ha) grid cells at the cost of shipping only. For selected comparatively flat topography, such as the Highveld (see Fig. areas in more remote regions DEMs of somewhat lower resolu- 2). However, such an interval is commonly insufficient to cap- tion (200 m x 200 m – 4 ha and 400 m x 400 m grid size – ture the required degree of detail in smaller areas, especially 16 ha) are available (CDSM, 2007).

54 Available on website http://www.wrc.org.za ISSN 0378-4738 (Print) = Water SA Vol. 36 No. 1 January 2010 ISSN 1816-7950 (On-line) = Water SA Vol. 36 No. 1 January 2010 The SRTM was an initiative, by the National Aeronautics

Gerhard Minnebron N and Space Administration of the USA (NASA), the USA’s Karst spring National Geospatial-Intelligence Agency, the German Space Agency and the Italian Space Agency, to produce a standard- ised, complete and high-resolution DEM of planet Earth using

Gerhard Minnebron Interferometric Synthetic Aperture Radar (InSAR) technol- wetland ogy (Farr et al., 2007). Taking 10 d or 149 orbits the mission covered an area between 60° N latitude and 56° S latitude leaving the largely uninhabited polar regions out. Each orbit produced a strip of images in 225 km wide swaths (Farr et al., 2007). Almost the entire targeted area (99.96%) was mapped at least once, with only a few patches over North America having been missed. Data errors of this mission are similar to those of Elevation of cursor position [m above mean sea level]: 1401m the USA’s National Elevation Dataset (Farr et al., 2007). With Cursor position InSAR cells generated at 1 arc-second by 1-arc-second (~ 30 m

Coordinates of cursor position x 30 m) the vertical height error for absolute error (i.e. meas- [degree minute second] Date of image [MM DD YYYY] ured against ground truth) is less than 16 m and the relative (i.e. measured against generated data) linear vertical height error is less than 10 m (Farr et al., 2007). Regarding the circular and horizontal geo-location, SRTM data have an absolute error of Figure 3 A screenshot of Google Earth™ showing the study area in 3-D less than 20 m and a relative error of less than 15 m. It should fashion with status bar at the bottom indicating elevation and also be noted that this method is not able to penetrate the veg- coordinate reading for the cursor position on the image etation canopy and thus, in densely-vegetated areas, overesti- mates ground elevation by the height of trees (Farr et al., 2007). Since the semi-arid Highveld, into which the study area falls, is However, for the vast majority of areas such informa- only sparsely vegetated with bushes and scrubs this error is of tion is not available, as was the case with the study area. This marginal, if any, importance. paper presents a cost-efficient alternative for creating the Since SRTM data underlie the Google EarthTM DEM, required level of vertical resolution by using satellite imagery attempts were made to use these data directly for generating provided, at a relatively low cost via the internet, by Google’s a high-resolution relief model of the wetland. However, the Google Earth™ application. The core of the method is based contours derived directly from SRTM data (as downloaded on elevation readings given by Google Earth™ for any posi- free of charge from the website of the Consortium for Spatial tion of a cursor which is moved across a chosen satellite image. Information – an initiative from the Consultative Group for Depending on user preference this elevation is provided in feet International Agriculture: http://srtm.csi.cgiar.org/), by sub or meters above mean sea level (m a. m. s. l.) at intervals of jugating them to high resolution interpolation, yielded unsat- 1 ft (approximately 30.5 cm) or 1 m, respectively, and displayed isfactory results with many contours showing an unnatural, in a control bar at the bottom of the image. Together with the blocky appearance and frequently being open-ended (Fig. 4, elevation, the exact coordinates for the locations indicated by next page). the cursor position are provided (Fig. 3). A possible reason for the fact that contours based on the In the specific case of this study it means that Google original SRTM data are less satisfactory than those generated Earth Google EarthTM’s elevation data are at a resolution 5 to 20 via Google EarthTM may be related to the continuous refinement times higher than available datasets provided by the CDSM. In of elevation data in Google EarthTM through successive addi- Google Earth Google EarthTM this dataset supports an additional tion of high resolution data from other sources as they become function which allows users to view the relief in their area of available for different areas (Taylor, 2009). This is implemented interest in 3-D fashion with a maximum vertical exaggeration by not only utilising satellites but also other (unspecified) factor of 3. For some areas in the GMB wetland this exaggera- technologies (Aubin, 2009). tion was insufficient to visualise height differences important Another reason may relate to the fact that SRTM-data for determining pathways of surface flow during flood events, had to be cut out from a larger set covering a 5° latitude by for example. However, once these data have been imported into 5° longitude grid cell since the computational power required a GIS the exaggeration factor can be chosen at will. to generate 1 m contours for the entire grid cell exceeded the In order to use this detailed relief information provided capacities of available computers. However, extracting the data by Google EarthTM in a DEM a method had to be designed relevant for the study area results in some deviation from the to import Google elevation data into a usable GIS format for main dataset owing to interpolation-related errors associated use in ArcDesktop™. This method is presented in this paper, with the newly defined margins of the cut-out area. This result- accompanied by a critical evaluation of the quality and reliabi ing misfit between original data from the entire grid cell and lity of the resulting GIS-based DEM. cut-out data for the study area is illustrated in Fig. 5 using 20 m intervals as an example (instead of 1 m intervals to reduce the Origin and accuracy of Google EarthTM elevation computational requirements). data In view of the above-described difficulties the extraction of improved elevation data as used in the final Google EarthTM Google Earth™ provides high-resolution elevation data using product was attempted. In addition to improved accuracy the virtual globe system, which started in June 2005 and used compared to SRTM data this had the additional advantage Shuttle Radar Topography Mission (SRTM) data for its eleva- of the relief information being geographically linked to tion baseline. continuously-updated land use data. Since the inception of the

Available on website http://www.wrc.org.za 55 ISSN 0378-4738 (Print) = Water SA Vol. 36 No. 1 January 2010 ISSN 1816-7950 (On-line) = Water SA Vol. 36 No. 1 January 2010 27°7'30"E 27°8'0"E 27°8'30"E 27°9'0"E 27°9'30"E

Legend Du Non-perennial stream 5 m SRTM contours for the study area T oi Perennial stream 1 m SRTM contours for the study area t S pr  u Wetlands (For the elevation ≤ 1405 m a.m.s.l) it r e 0 250 500 1.000 Meters v i Figure 4 (right) R i o 1 m interval contours of o M GMB (the wetland) as 26°28'30"S derived from SRTM data displaying an overall unnatural and blocky appearance

Figure 5 (bottom) Deviation between the 1400 m contour interval 26°29'0"S generated based on original SRTM data as contained in 5° grid cells and SRTM data extracted from the dataset for the study area (indicated by

double pointed arrows). 26°29'30"S Both contour sets not only display a blocky and unnatural appearance but also differ from the corresponding contour generated using elevation data extracted from 26°30'0"S Google Earth

27°7'30"E 27°8'0"E 27°8'30"E Legend peat project in late 2005 the imagery covering the 20 m SRTM contours for the larger study area wetland was updated 3 times, which is in line with 20 m SRTM contours for an entire tile Google Earth derived contours the average update interval indicated by Google Non-perennial stream EarthTM (Fig. 6, next page). Perennial stream Compared to available topographic informa- Wetlands ´ tion on the area, which dates back to 1975

Deviations between original (1:10 000 black and white orthophotographs) and and cut-out SRTM contours 1995 respectively (1:50 000 map series, sheet Rysmierbult 2627 AC), Google EarthTM imagery 26°29'5"S is significantly more recent. This is of particular importance with regard to possible quantification

26°28'55"S of impacts associated with peat mining, which results in a visible reduction of wetland areas and increasing extension of open water surface. With

0 most of the peat mining activities commencing on

4 1 a large scale after 2003, these impacts on the area’s

26°29'15"S hydrography are not captured in any of the standard topographic sources. However, owing to a rather low resolution of the first set of Google imagery for the study area, many of the required details were still obscured. Resolution improved significantly with the st1

26°29'55"S update covering the whole area of interest by

26°29'25"S images dated August - September 2004 (Fig. 6).

0 At this resolution, details such as the size and

4 1 shape of the several ponds created by peat mining within the wetland, the course of the drainage line, and the flow status of nearby streams and dams were discernable. The situation improved further with the 2nd update of imagery, after which the majority of the study area was now 0 62,5 125 250 Meters 26°29'35"S 0 250 500 1.000 Meters covered by satellite imagery as recent as June

27°8'5"E 27°8'15"E 2007 (Fig. 6, next page).

56 Available on website http://www.wrc.org.za ISSN 0378-4738 (Print) = Water SA Vol. 36 No. 1 January 2010 ISSN 1816-7950 (On-line) = Water SA Vol. 36 No. 1 January 2010

r e

Pre-2004 r August 2004 June 2007

i r

R S

F R S

o W

i F

o W

o M

W o

M M

open water open water open water Gerhard Gerhard Minnebron Gerhard Minnebron Minnebron

Figure 6 Sequence of Google EarthTM images of the wider study area, stretching from approximately 7 km upstream of the confluence of the Mooi River and the Wonderfonteinspruit (WFS), as updated between 2006 and 2008, displaying dates when underlying satellite imagery was captured. (Owing to increasing resolution in the more recent images a higher eye height had to be chosen in order to keep the size of the displayed area constant)

Generating high-resolution DEM using Google contour map underlying the DEM were assessed by com- Earth™ elevation data parison with standard CDSM contour data as well as high-resolution relief data generated from airborne survey The generation of a high-resolution DEM consisted of 3 general data. With an elevation interval of 0.5 m the latter provides steps: the highest resolution of all datasets and was therefore • In the 1st step, a procedure was developed to capture Google used as a benchmark. However, having been generated by Earth™ elevation point data and subsequently import them Computer Aided Design (CAD) software, the data first had into the GIS ArcDesktop™. to be converted into a GIS-compatible format. • The 2nd step comprised of the generation of a high-resolution contour map and DEM in GIS ArcDesktop™ using the A schematic overview of the process of generating a high- imported elevation data from Google Earth™. resolution DEM, in the form of a flow diagram, is depicted in • In the final step, the quality and reliability of the generated Fig. 7.

SRTM elevation data Google Earth image of study area with Elevation data

from other processing elevation data at 1m/ 1ft resolution Google intren intren Google sources

Grid-based capturing of Google Earth elevation data at 1m intervals

Export of captured data from Excel to GIS  Import of unprocessed (Arc Map) Generation of SRTM data of 5°-grid cell contour map for study area of 1 m into GIS Arc Map resolution (‚1m Google‘)

Quality assessment: Inter-comparison of generated ‚1m Google‘ map with other contour sets

Cutting out of CDSM contours in Airborne survey of study SRTM data topographic map area (CAD – format) relevant for study area 1:50000 of study area

Conversion of CAD data into GIS format

Generation of 5 m interval 20 m interval Generation of 1 m – interval contour (digital format only) (‚20m CDSM‘) 0,5 m – interval contour map of study area (‚5m CDSM‘) map of study area (‚1m SRTM‘) (‚0,5m Survey‘) Hardcopy Digital (jpg (vector format) format)

Figure 7 Flow path diagram depicting the data sources and different steps of generating a high resolution DEM/contour map based on Google EarthTM elevation data and its subsequent comparison to other contour datasets as a form of quality assessment (shaded blocks: contour sets that were compared with each other). For results of the inter-comparisons of the 6 different contour datasets see Table 2

Available on website http://www.wrc.org.za 57 ISSN 0378-4738 (Print) = Water SA Vol. 36 No. 1 January 2010 ISSN 1816-7950 (On-line) = Water SA Vol. 36 No. 1 January 2010 capital letters in alphabetical order from west to east, starting 27°7’22“ 27°9’27“ 3460 m at the column intersecting the confluence between the Gerhard

26°27‘48“ Minnebron stream and the Mooi River. Later 4 columns were 8“ added to the west named W to Z. Together with the rows num- 6“ bered 0 to 19 from south to north each point of the grid received N thus a unique alpha-numerical ID-code (Fig. 8). Each grid point was marked by a so-called ‘placemarker’, a Google EarthTM func- r e iv Grid point: E10 R tion in which additional information can be recorded and dis- i o o M played on screen. Using this ‘placemarker’ function for each grid

m point, the elevation was displayed when the cursor was held over 0

GMB eye 9 5

3 the point in question and manually recorded. In addition to the elevation of the grid points, every point at which elevation changed (by 1 m) was logged and captured together with its coordinates in a Microsoft Excel™ spread- sheet. Changes in elevation between grid points were estab- lished by slowly moving the cursor along the longitudinal and the latitudinal grid lines until the elevation reading changed. GMB wetland This was frequently done through zooming in to obtain a more accurate spatial reference of the point in question. All elevation

3470 m points were placed between 2 adjacent grid points (horizon- tally, east-west, or vertically, north-south) and named by com- 26°29’45“ bining the code of the 2 points separated by an arrow indicating the direction of cursor movement. Requiring an estimated 30 h, Figure 8 a total of 2080 elevation points were captured covering an area Google Earth™ exported image displaying the extent of target area around the Gerhard Minnebron (GMB) wetland and grid for of some 12.44 km² (equalling some 2.5 man-hours per km² at which elevation data at 1 m interval were extracted a rate of some 70 points/h). At an hourly rate of ZAR300 this would result in ZAR700/km². After inclusion of proportional Import of Google elevation data costs for the internet connection and the licence fees for the GIS software, this is over 10 times more cost-efficient, for this As a 1st step the extent of the target area had to be determined, area, than the high-resolution, airborne survey. taking possible hydraulic links between the wetland and the Points where the elevation changed by 1 m in a W-E direc- adjacent stream systems, especially overland flow during high tion are grouped as ‘horizontals’ while points changing in a flow conditions, into account. This resulted in a target area N-S direction are termed ‘verticals’ (Table 1). exceeding the actual size of the Gerhard Minnebron wetland Cases where elevation changes of more than 1 m between (Fig. 8). 2 neighbouring grid points occurred resulted in 2 or more Using the ‘display grid option’ of Google Earth™ this target points having the same point name. However, since x and y area was overlain by a grid of markers, with 6″ longitudinal and 8″ coordinate values are provided for a point they cannot be con- TM latitudinal difference between points resulting in a total of fused. The coordinates obtained from Google Earth had to be 16 columns and 20 rows (Fig. 8). The columns were named with converted from the degree-minute-second format into decimal

Table 1 Example of Microsoft Excel™ table used for recording information associated with elevation points for the GMB study area as retrieved from Google Earth (total length: 2 080 rows) Elevation South East Point Name Point Type DeciDegSouth DeciDegEast [m [DDMMSS] [DDMMSS] a.m.s.l.] 1399 262945 270752 A_0 Grid points -26.49583333 27.1311111 1400 262939 270752 A_1 Grid points -26.49416667 27.1311111 1400 262933 270752 A_2 Grid points -26.49250000 27.1311111 1401 262927 270752 A_3 Grid points -26.49083333 27.1311111 1409 262842 270722 W_10 Grid points -26.47833333 27.1227778 1410 262836 270722 W_11 Grid points -26.47666667 27.1227778 1411 262830 270722 W_12 Grid points -26.47500000 27.1227778 1412 262824 270722 W_13 Grid points -26.47333333 27.1227778 1408 262818 270722 W_14 Grid points -26.47166667 27.1227778 1407 262812 270722 W_15 Grid points -26.47000000 27.1227778 1409 262806 270722 W_16 Grid points -26.46833333 27.1227778 1417 262800 270722 W_17 Grid points -26.46666667 27.1227778 1420 262754 270722 W_18 Grid points -26.46500000 27.1227778 1415 262748 270722 W_19 Grid points -26.46333333 27.1227778 1400 262945 270726 W_0 -> X_0 Horizontals -26.49583333 27.1238694 1398 262945 270745 Z_0 -> A_0 Horizontals -26.49583333 27.1291111

58 Available on website http://www.wrc.org.za ISSN 0378-4738 (Print) = Water SA Vol. 36 No. 1 January 2010 ISSN 1816-7950 (On-line) = Water SA Vol. 36 No. 1 January 2010 degrees to be usable in ArcView GIS. This conversion was (IDW) technique of ArcDesktop™ to interpolate between executed as follows: these points a 1 m interval contour map of the study area was generated (Fig. 9). a.. DeciDegEast = deg + (minute/ 60) + (second/ 3600) Based on the retrieved set of elevation data from Google b.. DeciDegSouth = - deg - (minute/ 60) - (second/ 3600) EarthTM a DEM was generated in ArcDesktop™ allowing for a 3-D-view of the study area. Enhancing elevation data through where: colour-coding aided the identification of potential pathways for DeciDegEast and DeciDegSouth are the decimal degree water from the Mooi River possibly entering the peatland dur- coordinates for the latitudes and longitudes, respectively. ing flood events as well as additional outflow (seep) points Newer versions of Google Earth™ tools/options show how (Fig. 10, next page). latitude/longitude can be changed, sparing the user from manually calculating these coordinates. Assessing quality of the generated contour map

It should be noted that the resolution could be increased by a In order to assess the reliability of the generated contour map factor of 3, to 0.3 m contour intervals, by choosing in Google and the associated DEM of the study area, the contour infor- EarthTM to display elevation in feet instead of meters. Since this mation was compared against 2 sets of standard topographic would have resulted in approximately triple the amount of time data (20 m and 5 m contour intervals from CDSM). The latter spent on capturing the elevation points, only changes at 1 m 2 sets were also compared to each other to assess how much intervals were used in this pilot study. same-source contour data may deviate from each other. This, in turn, provides some baseline as to what degree of deviance Generating contour maps and DEM in GIS is regarded as acceptable by the CDSM, and was thus used to ArcDesktop™ assess the quality of the generated 1 m Google contour data. In addition, elevation data from an aerial photography Out of the 2 080 points logged, only points where the eleva- survey later became available to the project. This dataset tion displayed in Google EarthTM changed, e.g. from 1 410 m provided the study with, among other features, 0.5 m con- to 1 409 m (in total 1 760 points, termed ‘horizontals’ or ’ver- toured elevation data (acquired digitally from DWAF, 2008). ticals’) were used to generate the contour map, resulting in a Having the highest resolution this dataset was used as a density of 1 elevation point per 7 069 m². The remainder are benchmark against which all other contour datasets were grid points that were omitted since their elevation could not compared. Generated in a computer-aided design (CAD) be determined as accurately, as most of them fall somewhere program, digital elevation data underlying this map had to in between points where elevation changes. The use of the 320 be converted to be used in the ArcView GIS. This procedure, grid points would therefore have introduced an error margin as well as pertinent technical details of the airborne survey, of up to 1 m per point. Using the Inverse Distance Weight is briefly outlined below.

27°7'20"E 27°7'40"E 27°8'0"E 27°8'20"E 27°8'40"E 27°9'0"E 27°9'20"E 1410

5 1 1

4 4 0 3

1 2

0 5 4

1 1

0 26°28'20"S 5

0 5 1 40 4 1 1 0 5 0 4 1 26°28'40"S

0 0 4 1

0 Original image in Google Earth displaying the grid

from which a total of 1 760 26°29'0"S 5

elevation points were retrived 3

1 4 underlying the map displayed 4 1 1 on the right hand side 0 5  0 4 0 1 3 4 Legend 1 1:20,000 0

4  26°29'20"S 0 Canal 4

0 1

4 1 1m-Google contour Non-perennial stream Perennial stream 00 14 0 250 500 1,000 m 5 2

4 Wetlands 26°29'40"S 1

Figure 9 Map of the study area displaying 1m interval contours generated through IDW (inverse distance weight) interpolation in GIS ArcDesktop™ based on 1 760 point elevation data retrieved from Google Earth™

Available on website http://www.wrc.org.za 59 ISSN 0378-4738 (Print) = Water SA Vol. 36 No. 1 January 2010 ISSN 1816-7950 (On-line) = Water SA Vol. 36 No. 1 January 2010 N

Possible inflow pathways during flood events in the Mooi River Figure 10 Colour-coded DEM (based M on IDW ooi R iver interpolation) of the wider Gerhard Minnebron wetland area Du Toit Spruit based on 1 m interval Google Earth™ elevation data (from the SRTM) showing possible event- driven flow paths between GMB wetland the Mooi River and the wetland

X +2 929 200 X +2 929 200

1 4

0 4

4 0

. 7

5 4

0 5 Legend 1 Y - 14 900 14 - Y Y - 14 800 14 - Y Y - 14 700 14 - Y Y - 14 600 14 - Y Y - 14 500 14 - Y Y - 14 400 14 - Y Y - 14 300 14 - Y Y - 14 200 14 - Y

Y - 14 100 14 - Y 5

4 4 0 5 . . 5 0 0 1 1406.0 5 4 . 0 5 7

1403 .0

. 0 140 3

1406.5

0 4 . 5

4 140

0 6

4 .0 .

0 1

4 1

0 0 1

3 4 6 .

5 05

. .

5 5

0 1

Rocky 3 . 0 4 d6.dgn Polyline 0 5 .0

Rocky 1

4 .

5  Fallow Land  .0

4 Cad Renderer 0 2

0 Ruin .

0 Ruin

1 .

4 Res 1:400,000,000

. 0

X +2 929 300 1 X +2 929 300

3 1

0 9 9 9

0 .

5 . . 0

. Ruin

9 8

. 5

3 Solid, 6, 0

9 W/Tank 5

8 6

. 0

0 4

0 4 2 4

7 . 0

. 5 Fallow Land Ruin

1405 Solid, 4, 2

Ruin Ruin

140 6 140 6 Solid, 2, 0

139 6 Solid, 0, 0

0 4 . 0 1 4 0 X +2 929 400 3 X +2 929 400 . 5

1 3 9 Medium-Dashed, 0, 0 7

. 0 .0 5

0 . 4

5 0

. 4

. 5 4

5 0 . 6 5 1

0 .

0 4 0

6 4

8 0

0 1 4 .

7 1

1 1

1 4 4

1 1

0 8 .

1396 139 8 5

6 .

5 1408 Solid, 1, 1 Figure 11

0 1

139 6 8 140 2 .

0 An example

1402 Solid, 1, 0

1 5

4 .

0 5 7

. 0 5

. 4

6 1

4 3

. 0

. 5

6 1

0 Land 6

1 for 1 out of

X +2 929 500 4 X +2 929 500

0 5

5 Short-dashed, 1, 0

. 0

. 5

1 1

4 4

0 0

. 0 .

5 140 6

1 22 separate 4

3 W/Tank . 5 5

1 0

4 1

0 4

4 3 1

0 0

1 3

4 LongDashed, 0, 1

0 2

5 1

2 Concr. .

0 1

4 4

1 0 .

5 2 1

. 140 5 4

5 0

1 6

4 .0 CAD design

1 . 0

1 05.

4 14 5

0 0

0 2.0

. 5 Land Short-dashed, 56, 0

140 6

1 6

3 0 1

4 0 9

0 14 0

9 05

. 2 .0

9 . 5 files covering

1404.5 .5 01 14 140 0 1 Solid, 0, 1 4 04 .0 an 800 m long Reeds 140 4

1.0

X +2 929 600 140 X +2 929 600 Rocky

0 .

9 3

1 00.5 1

3 14 9

3 8

8 .

5 0

. Solid, 1, 2 . 5 7 9 1 3

. 7 9

1 3

. and 500 m 6

9 3 0 140 3

1 .

0 1

3 0 9

6 . 0 4 1 Rocky

5 .

9 9

5 9 . 5 139 9 140 2 3 140 2 1 1403.0 .0 9 9 3 Short-dashed, 0, 0 1 wide part of

. 5 8 9 1396 3 1

0 . 1 8 4 9 0 3 3 . 1 5 5

9 0 3 . the surveyed

. Medium-Dashed, 6, 0

3 1

95 1398

. 0 6 9 1

3 study area

Y - 14 900 14 - Y Y - 14 800 14 - Y Y - 14 700 14 - Y Y - 14 600 14 - Y Y - 14 500 14 - Y Y - 14 400 14 - Y Y - 14 300 14 - Y Y - 14 200 14 - Y

Y - 14 100 14 - Y Short-dashed, 2, 0 X +2 929 700 X +2 929 700

o o CO-ORDINATE SYSTEM LO. CLARKE 188027 CO-ORDINATE SYSTEM LO. C LAR KE27 1880 SHEET LAYOUT DIRECTORATE SPATIAL AND LAND REPUBLIC OF SOUTH AFRICA INFORMATION MANAGEMENT R. L. FROM G.T.S. BEACONS 1492.80 METRES ABOVE M.S.L. R.L. FROM G.T.S. BEA CONS 1492.80 METRES ABOVE M.S.L. C5 C6 PRIVATE BAG X313 D5 D6 D7 PRETORIA DEPARTMENT OF WATER AFFAIRS & FORESTRY STATION Y X R.L. STATION Y X R.L. 0001 E4 E5 E6 E7 Dotted, 0, 0 F3 F4 F5 F6 F7 MOOI RIVER G.W.S G3 G4 G5 G6 SURVEYED Gauteng Region 01/'08 TRACED Siyaz i DTM Services 04/'08

H3 H4 H5 COMPILED Siyazi DTM Services 04/'08 TOPOGRAPHY CHECKED J4 J5 CONTOUR SURVEY OF GERHARDMINNEBRON PEAT STUDY

PHOTOGRAPHY W611 03/'08 CADASTRAL CHECKED

0 10 20 40 60 80 100 120 140 Dotted, 0, 1 No. REVISION DATE LOC. NO. REG. NO. m DISTRICT POTCHEFSTR OOM SHEET D6

SCALE 1/1000 AREA 40 Ha CODE SAR CON DSM C230/41 155335/08 Solid, 0, 2 Dotted, 0, 2 Aerial photography survey elevation map dated March 2008 Solid, 3, 2 Generated by Computer Aided Design (CAD) software (.dgn-file type) Solid, 8, 0 Source: DWAF (2008) Solid, 8, 2

Apart from the visual comparison employed in this Airborne relief mapping (0.5 m contour interval) paper there are other, more formalised and quanti- The mapping is based on a stereoscopic evaluation of a series fiable, procedures such as those used by Gorokhovich of aerial photographs taken from an aircraft at a height of 4 000 and Voustianiouk (2006) and Sun et al. (2003). However, m above ground level, accurate enough for use in 1:2 000-scale in view of the associated debate regarding the best plans (Smith, 2009). Twelve ground-control points of known method, and the fact that visual comparison proved to elevation and coordinates were employed to geo-reference the be sufficient for the purpose intended, no further efforts images acquired from the aircraft. Geo-referenced stereoscope were made to quantify deviations between the different images were produced and, by using these images and a stereo- contour sets in a formalised manner. plotter, contours and other features were drawn in (Smith, 2009).

60 Available on website http://www.wrc.org.za ISSN 0378-4738 (Print) = Water SA Vol. 36 No. 1 January 2010 ISSN 1816-7950 (On-line) = Water SA Vol. 36 No. 1 January 2010 27°7'40"E 27°8'0"E 27°8'20"E 27°8'40"E 27°9'0"E 0 250 500 1,000 m Figure 12 Legend

26°28'20"S Map with 0.5 m contours generated from the 0.5m-Survey contours in 1m intervals aerial photography survey data displaying Area covered by reeds reed cover and open water areas Area covered by open water Non-perennial stream Perennial stream 26°28'40"S  1:15,000 26°29'0"S

27°8'30"E 27°8'40"E 27°8'50"E

1397 7 0 14  26°29'20"S 6 1:2,500 0 4

26°28'40"S 1 405 1 1 4 0 4

14 03 1403

26°29'40"S 1402 01 1 14 400 9 1397 139 1396 1396 1398

6 139 26°30'0"S 1396

1395 26°28'50"S

Legend

0.5m-Survey contours in 1m intervals Area covered by reeds 1397 0 62.5 125 250 m Area covered by open water

The data from the airborne survey were then digitised data and mapped topographic features was detected; these using a CAD program locking all feature information in a misfits are possibly related to inaccuracies in geo-referencing single CAD file (design files, .dgn). Within these files colour and projection. However, for the purpose of the project, the and cartographic symbols are used to identify different fea- misfits were regarded as tolerable. The DEM produced from tures. An example of such a file for a segment of the study area the CAD dataset differs from the Google-based DEM in that it is displayed in Fig. 11. was based on CAD-generated contours and not on interpolated The conversion of CAD data into GIS-compatible shape files elevation points. Figure 13 shows the CAD-based DEM of the was complicated by the fact that for symbols consisting of more GMB wetland also displaying some of the features captured by than 1 part (such as a dashed line which is made up of many the aerial survey. small lines), each separate part of that symbol is identified in GIS as a topographic feature. While these single parts are still automatically grouped N into one larger shape file (e.g. a catchment bound- ary), such shape files cannot be used for determin- Structures GMB ing enclosed areas. This is only possible once Rock outcrop Wetland each part of the shape file is connected to others Bushes forming one larger entity (using among other Contour lines techniques the ’merge function’ in the ‘Editor’ tool of ArcDesktop™). Where CAD lines were enclos- ing areas such as mining ponds, wetlands, etc., of which the area size had to be determined, this connection of separate symbol parts had to be per- C an formed, rendering the import of CAD data rather al time-consuming. Before the area enclosed by such Mo oi Rive lines could be determined, all these lines had to r be linked together to form an uninterrupted line Farmland around the area which could then be transformed into a polygon using the ‘buffer’ and ‘union’ functions in ArcDesktop™. Only after the poly- gons are created, can the size of enclosed areas be 500 m determined. Examples for such areas (wetlands and mining ponds) are displayed in Fig. 12. Figure 13 Compared to the 1:50 000 topographic map, DEM based on aerial photography survey contour data (0.5 m intervals) which spatial deviation between the GIS-imported CAD include features such as rock outcrops, structures, vegetation, rivers and canals

Available on website http://www.wrc.org.za 61 ISSN 0378-4738 (Print) = Water SA Vol. 36 No. 1 January 2010 ISSN 1816-7950 (On-line) = Water SA Vol. 36 No. 1 January 2010 27°8'30"E 27°9'0"E 27°9'30"E 27°10'0"E  0 250 500 1,000 m 0 250 500 m 1:20,000  1:5,000

 26°28'10"S

26°28'0"S

4 1 Figure 14 Mismatch between same-source data:

26°28'30"S ‘Original contours’ 26°28'20"S being the contours 1440 before mismatch was reset and ‘CDSM fitted contours’ being the contours set to the

26°29'0"S form and location of the 27°9'40"E 27°9'50"E contours found on the national topographic maps (1:50 000) Legend

Fitted countours Original contours 26°29'30"S Non-perennial stream Perennial stream

27°7'0"E 27°7'30"E 27°8'0"E 27°8'30"E 27°9'0"E 27°9'30"E were subsequently taken as the base it u r p dataset to adjust the digital (vector) Mo s oi in R e iv t contour lines (Fig. 14). er n o f r e The adjusted set of the digital d n o W 20 m interval contour data was

26°28'0"S subsequently used for the following

comparisons.

4 3

8 Max. horizontal deviation

Du m To it Sp Deviations between same- ru it source data: 20 m CDSM vs. 5 m CDSM contours 26°28'30"S Owing to the generally flat topo graphy of the wetland only 3 x 20 m interval contours occur within 1400 the larger study area and can thus be compared to the 5 m interval 1400 contours. Figure 15 depicts the 26°29'0"S

 1420 deviations between 2 sets of con-

1400 tour lines both originating from the 1400 1420 CDSM. Legend In assessing the deviation Non-perennial stream 1440 between contour lines, it is impor- Perennial stream tant to note that identical degrees of

26°29'30"S 5m-CDSM contours 1440 20m-CDSM contours deviation in steep terrain result in Wetlands generally smaller horizontal dis- 0 250 500 1,000 m tances between contour lines (i.e. ‘misfits’) than in flat terrain where Figure 15 the lower topographic gradient (the Deviation areas between 20 m and 5 m CDSM contours illustrated using the 1 400, height difference kept constant) 1 420 and 1 440 m a. m. s. l. contour lines of both sets results in a larger horizontal distance between contour lines. With Comparing Google and CDSM data some 438 m the largest horizontal misfit occurs at the 1400 m Before comparing 20 m interval CDSM contour data with contour in the (flat) confluence area where the Du Toit Spruit the 5 m CDSM contours it was noted that the contour data in joins the Mooi River (Fig. 15). vector format (i.e. discrete .shp files) showed some deviation from identical contours as displayed in scanned images of 1 m Google contours vs. 20 m CDSM contour data 1:50 000 topographic maps imported into ArcDesktop™. The To compare the 1 m contour map (‘Google contours’), only the topographic image’s contour lines (as the original data source) 2 x 20 m CDSM contour lines (‘CDSM contours’) which occur

62 Available on website http://www.wrc.org.za ISSN 0378-4738 (Print) = Water SA Vol. 36 No. 1 January 2010 ISSN 1816-7950 (On-line) = Water SA Vol. 36 No. 1 January 2010 in the area of interest could be used, 27°7'0"E 27°7'30"E 27°8'0"E 27°8'30"E 27°9'0"E

it u namely the 1 400 m and the 1 420 m r M p oo s i n R i a. m. s. l. lines. Comparing these with iv e er t n o f the corresponding contour lines, from r e d n the generated Google map, suggest an o W

overall satisfactory fit between the 26°28'0"S 2 sets of contours (Fig. 16). However, Google contour lines Max D . hor u izont matched best with CDSM contours To al de it 540 m viatio Sp n which generally displayed a 5 m ru it lower elevation (Fig. 16). The reason for this 5 m off-set could not be deter- r

26°28'30"S r u u o mined. It is, however, relatively easy o t t n n o to adjust by subtracting 5 m from o c c 0 5 0 all contour heights in the generated 0 4 5 4 1 0 1 4 Google map. As was the case with M 1 le S 0 0 g D 0 0 the 5 m vs. 20 m CDSM contour data, o 4 4 C 5 1 o 1 s 0 G t i 4

the largest horizontal deviation f 1 5 (540 m) occurs at the 1 400 m con- 26°29'0"S 0 4  1 tour at the confluence area of WFS and Mooi River, this time, however,

5 r extending into the Du Toit Spruit area 2 05 r u Legend 14 4 u o 1 o t (Fig. 15). Based on the maximum hor- t n n Non-perennial stream 0 o o c 0 0 2 izontal deviation it appears that the 0 c Perennial stream 4 0 4 0 1 4 1 5 2 1 2 4 generated 1 m-Google map fits well 4 1 26°29'30"S 20m-CDSM Countours 1 M e to the 20 m CDMS contours (with a 1m-Google contours l S 0 g D 2 o C maximum horizontal deviation of 4 o Wetlands 1 G s it 540 m) while the 5-m-contour set fits 0 250 500 1,000 m f to the 20 m CDMS contours (also created by the CDSM) with a maximum Figure 16 horizontal deviation of 438 m. The 1 m Google contours compared with 20 m CDSM contours illustrating an overall good fit fact that the generated map based on and a +5 m off-set in Google data using the 1 400 m and 1 420 m a. m. s. l. contour lines Google data displays an even better indicating a maximum horizontal misfit of 540 m between the 2 sets of contour lines fit to the standard 20 m CDSM contour than the 5 m contour generated 27°7'0"E 27°7'30"E 27°8'0"E 27°8'30"E 27°9'0"E 27°9'30"E it u r by the CDSM themselves indicates a p Mo s oi in R e better-than-expected reliability of the ive t r n o f r e generated Google map. d n o W 1-m-Google contours vs. 26°28'0"S 5-m-CDSM contour data Max. horizontal deviation Comparing 1 m Google contours Du T oi 490 t S m with 5-m-CDSM contours once again pr ui showed an overall good fit between t the 2 datasets (Fig. 17).

26°28'30"S 5 As observed in the 2 previous 140 5 0 4 comparisons the largest horizontal 1 00 deviation occurs again on the 1 400 m 14 contour line next to the confluence of the Mooi River and the Du Toit Spruit

(490 m) (Fig. 17). As demonstrated 0 26°29'0"S 40 1 5 earlier, this is most likely not due to  0 4  1 an above average error in either of the datasets but rather due to the fact that 5 0 Legend 4 this area is particularly flat resulting 1 Non-perennial stream 5 2 in larger horizontal misfits than iden- 4 Perennial stream 1 tical misfits in steeper terrain. 20 26°29'30"S 1m-Google contour 0 14 41 5m-CDSM contours 1 0 5 3 1 4 Comparison of all contour data Wetlands 4 1 5 1 3 0 250 500 1,000 m 4 to a high-resolution airborne 1 survey: 0.5-m-contour interval The survey was done at 2 different Figure 17 levels of accuracy with the GMB Selected 1 m Google contours (bold dashed line) compared against corresponding wetland itself mapped at a scale of 5 m CDSM contours (5 m off-set to be considered)

Available on website http://www.wrc.org.za 63 ISSN 0378-4738 (Print) = Water SA Vol. 36 No. 1 January 2010 ISSN 1816-7950 (On-line) = Water SA Vol. 36 No. 1 January 2010 1: 4 000 while the wider study area extending to the conflu- For the purpose of comparing contour lines only the area ence of the WFS and the Mooi River was captured at a scale of mapped at higher accuracy (1: 4 000) covering the actual wet- 1:8 000. However, the contour interval resolution for both areas land is used. The 0.5 m contour interval map produced for the was kept at 0.5 m. The location of the 2 areas mapped using study area is displayed in Fig. 19. airborne survey data is indicated in Fig. 18. The noticeable lack of contour lines in the actual wetland illustrates the flatness of the area, where differences in the micro-relief driving the development of flow patterns are frequently well below the 0.5 m interval. It should also be noted that the confluence area of the Du Toit Spruit and the Mooi River falls outside the 0.5 m part of the survey (Fig. 19). For the subsequent comparison with contour data of alternative origin, only those contour lines of the 0.5 m set which correspond with the contours of the different datasets falling into the surveyed area are displayed.

20 m CDSM contours: Measured against 2 contours from the 20 m CDSM contour dataset (1 420 m and 1 440 m a. m. s. l.) the aerial photography survey contours require no off-set correction displaying an overall good degree of congruence between the 2 contour sets (Fig. 20). Owing to the limited area for which the survey at this resolution was done, only parts of the 20 m contours can be compared against corresponding 0.5 m survey contours. For the area covered, a comparatively small horizontal deviation of a maximum 150 m at the 1 400 m contour line indicates an overall good fit between the 2 datasets (Fig. 20, next page). Figure 18 Topographic map of the study area depicting the areas surveyed 5 m CDSM contours: As Fig. 21 (next page) indicates, the by airborne photography using 2 different scales of mapping 0.5 m survey contours and the 3 relevant 5 m CDSM contours (1:4 000 for the wetland – Area A and 1:8 000 for the wider study (i.e. 1 400 m, 1 405 m, 1 410 m) match well. area, Area B) (DWAF, 2008)

27°7'0"E 27°7'30"E 27°8'0"E 27°8'30"E 27°9'0"E

it u 4 14 r 40 1 M 1 2 o p oi s R in iv e er t n fo r

26°28'40"S e d 1401 n o W 26°28'0"S

1398

Du To it Sp ru it 26°28'30"S

1401 26°28'50"S

1404 0 6 2 1 4 2 4 1 1 1 4 1 26°29'0"S

8 0 4 1

27°8'50"E 27°9'0"E 27°9'10"E Legend Non-perennial stream Canal Perennial stream Wetlands  0.5m-Survey contours 025 50 100 m 26°29'30"S 1:2,500 Legend Non-perennial stream Perennial stream 0.5m-Survey contours 1m-Google contour Figure 19  Wetlands 0.5 m survey contours for the wider study 0 250 500 1,000 m area as well as the wetland part 26°30'0"S

64 Available on website http://www.wrc.org.za ISSN 0378-4738 (Print) = Water SA Vol. 36 No. 1 January 2010 ISSN 1816-7950 (On-line) = Water SA Vol. 36 No. 1 January 2010 27°7'0"E 27°7'30"E 27°8'0"E 27°8'30"E 27°9'0"E 27°9'30"E

it u r p Mo s oi in R e iv t er n o f r e d n o W 26°28'0"S

Du To it Max. horizontal deviation Sp ru it 150 m

1420 26°28'30"S Figure 20 1400 20 m CDSM contours compared against 0.5 m 1400

survey contours 1420

1400 26°29'0"S  0 0 1400 4 1

Legend 1400 Non-perennial stream Perennial stream

26°29'30"S 20m-CDSM contours 0.5m-Survey contours Wetlands 0 250 500 1,000 m

27°7'0"E 27°7'30"E 27°8'0"E 27°8'30"E 27°9'0"E 27°9'30"E

it u r p Mo s oi in R e iv t er n o f r e d n o W 26°28'0"S

Du To it Sp ru it 1405

Figure 21 26°28'30"S 1405 5 m CDSM contours as solid lines and the 0.5 m 1400 1405 survey contours as broken lines with each contour identified by a unique 1400 signature 1400

26°29'0"S 0 0 4 1  0  1 5 4 0 0 1 0 4 4 1 0 1 0 4 1 Legend 140 m 1405 Max. horizontal deviation Non-perennial stream

0 Perennial stream 1 4 1

26°29'30"S 0.5m-Survey contours 5m-CDSM contours Wetlands 0 250 500 1,000 m

Unlike deviations in previous comparisons, the maximum elevation for each contour line by 5 m. Owing to the relatively deviation was not found to be associated with the lowest-lying large number of contours only the following 5 pairs of corre- contour line in the study area (1 400 m a. m. s. l.) but occurred at sponding contour lines are compared with each other: 1 400 m, the 1 405 m contour where it amounts to some 140 m (Fig. 21). 1 405 m, 1 410 m, 1 415 m and 1 420 m a. m. s. l. (Fig. 22). The deviation between the 0.5 m survey and 1 m Google 1 m Google contours: Before comparing the 2 datasets the contours appears to be more complicated than a simple misfit, Google contours had to be corrected by reducing the displayed with a maximum horizontal deviation of 221 m (Fig. 22).

Available on website http://www.wrc.org.za 65 ISSN 0378-4738 (Print) = Water SA Vol. 36 No. 1 January 2010 ISSN 1816-7950 (On-line) = Water SA Vol. 36 No. 1 January 2010 In addition there is also a sporadic difference between the Conclusions and recommendations shape of 1 m and 0.5 m contour lines covering the study area. The results of the various inter-comparisons of different A common problem for hydrological research in topographi- contour sets discussed above are summarised in Table 2. cally flat areas such as wetlands is the lack of sufficiently

27°9'0"E 26°28'30"S

5 1 0 4

4 1

27°7'0"E 27°7'30"E 27°8'0"E 27°8'30"E 27°9'0"E 27°9'30"E

Max. horizontal deviation

2 2 26°28'0"S

Du To it S pr uit 26°28'30"S

05 14 0 0 4 1 1400

0 0 4 1 26°29'0"S 0 0 0 4 0  1 1 2 4 1 4 1 5 0 1 4 40 1 Legend m  Non-perennial stream Perennial stream

26°29'30"S 0.5m-Survey contours 1m-Google contour Legend Wetlands 0 250 500 1,000 m 0 Non-perennial stream 1 4

5 1

Perennial stream 0

4 1m-Google contours 1 1400 Figure 22 0.5m-Survey contours 5 2 1 m Google contours (dashed) Wetlands 1405 4 5 0 1 1 0 62.5 125 250 m 1 and 0.5 m survey contours (solid) 4 4 1 1 compared for selected contours 26°29'0"S

Table 2 Overview of results of the inter-comparison of different contour datasets available for the study area with the generated 1m Google contour map 1 m SRTM 5 m CDSM 20 m CDSM 0.5 m survey (vector shp) (image – jpg) Figs. 5 and 6: Fig. 17: Fig. 16: Fig. 22: - SRTM contours are blocky - A vertical off-set of - 2 x contour lines evaluated - A vertical off-set of 5 m was detected and often open-ended 5 m was detected (1 400/1 420 m) - More complex pattern of deviation: - Horizontal misfit of approx. - After correcting this - A vertical off-set of 5 m was detected  horizontal misfit max. 221 m 50 m between original 5° x off-set an overall - After correcting this off-set an over- (1 405 m contour line) 5° grid cell data and those cut good fit appeared all good fit appeared shape of contours out for study area - Max. horizontal - Maximal horizontal deviation 540 m - Deviations mainly confined to flat - 1 400 m contour is split in 3 deviation: 490 m (1 400 m contour line) areas; in steep terrain generally good 1 m Google parts and deviates from cor- (1 400 m contour - 1 m Google contours appear some- fits responding Google contour line) what smoother than 20 m CDSM contours Fig. 15: Fig. 21: - 3 x contour lines evaluated (1 400/ - 3 x contour lines compared (1 400/ 1 420/1 440 m) 1 405/1 410 m) - Maximal horizontal deviation 438 m - An overall good fit - - (1 400 m contour line) - Max. horizontal deviation: 140 m - Positive and negative deviations (1 405 m contour line)

5 m CDSM occur (vector – shp) - 5 m contour line appears to be smoother than 20 m line Fig. 14: Fig. 20: - At 1:50 000 map for study area a - only 2 x 20 m contour lines could be horizontal misfit of approx. 30 m compared - - was detected (1 400 m contour line) - Overall good fit - Positive and negative deviations - Maximal horizontal deviation: 150 m

20 m CDSM occur (1 400 m contour line) (lvector - shp)

66 Available on website http://www.wrc.org.za ISSN 0378-4738 (Print) = Water SA Vol. 36 No. 1 January 2010 ISSN 1816-7950 (On-line) = Water SA Vol. 36 No. 1 January 2010 detailed elevation data required to reliably delineate catch- in the studied wetland which despite numerous field visits ment areas, understand flow patterns within the wetland and remained invisible to the naked eye. This, in turn, aids the to identify potential hydraulic interactions between wetland understanding of how landforms may influence hydrologi- and adjacent streams, to name but a few examples. GIS-based cal processes such as direction of water flow, development DEMs prove to be useful tools in addressing most questions. of drainage patterns, occurrence of groundwater discharge Compared to standard topographic maps displaying 20 m areas, etc. Applying the computer-generated DEM to the contour intervals, and 5 m intervals for selected areas, eleva- study area also aided the delineation of catchment boundaries tion data provided by Google EarthTM, an internet-based global and identification of presently unmapped surface depressions visualisation application, are provided at a high enough resolu- along which flood waters enter the wetland. Moreover, maxi- tion to create 1 m intervals, at a minimal cost. mum peat accumulation levels could also be determined as Although SRTM data underlie the Google Earth™ relief they are controlled by the elevation of the water table at the information, the direct use of freely-available SRTM data outflow point. downloaded from the internet yielded unsatisfactory results In summary it is concluded that the presented methodology that were found to be inferior to contours generated by the allows the retrieval, conversion and application of elevation procedure described in this paper. Apart from errors associ- data from Google Earth™ in order to produce reliable GIS- ated with the interpolation of elevation data cut-out from based DEMs that have significantly higher resolutions than the larger sets, this may be caused by the continuous improve- standard topographic maps. These DEMs and contour maps ment of the terrain database in Google Earth™ which in have proven to be particularly useful tools for hydrological addition to satellites also includes other technologies and research in low-relief energy environments such as wetlands alternative data sources. This may partly explain why the for which adequate contour data are either not available, or too vertical accuracy of Google elevation data is higher than costly to generate. the 16 m indicated for the original SRTM data. Even after introducing potential errors associated with the extraction of Acknowledgements elevation data described in this paper, the generated contours and DEMs at 1 m contour resolution were found to compare Funding of the project (Project No. 2006-231) and logistical favourably with not only relief data of standard topographic support by DWAF (now DWA) is gratefully acknowledged. maps but also results of a high resolution aerial survey. The researchers would also like to give heartfelt thanks to However, with little or no information available on the exact Ms S Rudolph and the DWA agents at the Boskop Dam and procedures and interpolation techniques used in Google Aquapark offices for their patience, help, data and other saving EarthTM that so obviously improve the original SRTM dataset, graces, as well as to T de Klerk and D Cilliers for their input. such good agreement may be area-specific and can therefore We also wish to thank an anonymous reviewer, in particular, not be unconditionally extrapolated to other regions. for her/his encouraging and constructive comments. In addition to higher resolution relief data, Google Earth™ also provides more up-to-date hydrographical information, as References a result of the close updating intervals – 2 to 4 yr compared to the approximately 20 yr intervals for the 1:50 000 topographic ALEMAW BF and CHAOKA TR (2003) A continental scale water maps. This is of particular importance in environments where balance model: a GIS-approach for Southern Africa. Phys. Chem. anthropogenic impacts such as peat mining result in fast- Earth 28 957-966. AL-SABHAN W, MULLIGAN M and BLACKBURN GA (2003) A changing hydrographical conditions. real-time hydrological model for flood prediction using GIS and the The method developed to capture elevation data from WWW. Comput. Environ. Urban Syst. 27 9-32. Google Earth™ and import these into ArcDesktop GIS proved AUBIN M (2009) Google Earth: From Space to Your Face…and to be successful in creating a reliable, high-resolution DEM Beyond. URL: http://www.google.com (Accessed 17 August 2009). for the study area. After the off-set corrections of 5 m, devia- BOBBA AG, SINGH VP and BENGTSSON L (2000) Application of tions of the generated 1 m contour lines from 20 m contours environmental models to different hydrological systems. Ecol. contained in the 1:50 000 topographical map of the CDSM, the Modell. 125 15-49. deviations were found to be less than those between the CDSM (2007) Personal communication. 28 June 2007. Enquiry 4253/07. E-mail from [email protected] to E Hoffmann. 2 CDSM contour datasets (i.e. 20 m and 5 m interval con- DROGUE G, PFISTER L, LEVIANDIER T, HUMBERT J, tours). As sources for potential errors it should be noted that, HOFFMANN L, EL IDRISSI A and IFFLY JF (2002) Using 3D apart from some deviations between the 20 m and 5 m CDSM dynamic cartography and hydrological modelling for linear stream- contour lines, we also detected discrepancies between printed flow mapping. Comput. Geosci. 28 928-994. contours in scanned images and contours in digital format DWAF (2008) Topographic map 1:50 000 sheet 2627 AC (vector data), both provided by the CDSM. (Rysmierbult), 1995 indicating the flight direction and number of Compared to contour lines at 0.5 m intervals from a photo- strips of the airborne photography survey. Department of Water grammetric airborne survey, more complex patterns of devia- Affairs and Forestry, Pretoria, South Africa. FARR TG, ROSEN PA, CARO E, CRIPPEN R, DUREN R, tion to the generated 1 m Google contours were found, while HENSLEY S, KOBRICK M, PALLER M, RODRIGUEZ E, ROTH the (significantly less dense) CDSM contour data continued to L, SEAL D, SHAFFER S, SHIMADA J and UMLAND J (2007) show an overall satisfactory degree of congruence. However, The Shuttle Radar Topography Mission. URL: http://www2.jpl. with approximately ZAR10 000 per km2 for the airborne nasa.gov (Accessed 30 June 2008). DOI:10.1029/2005RG000183. survey the presented method of sourcing elevation data from FRASER CJD, ROULET NT, and LAFLEUR M (2001) Groundwater Google EarthTM is over 10 times more cost-efficient, provided flow patterns in a large peatland. J. Hydrol. 246 142-154. that internet access, appropriate GIS software and entry-level GOROKHOVICH Y and VOUSTIANIOUK A (2006) Accuracy GIS operating skills are available. assessment of the processed SRTM-based elevation data by CGIAR using field data from USA and Thailand and its relation to The generated 1m resolution DEM was found to signifi- the terrain characteristics. Remote Sensing Environ. 104 409-415. cantly enhance the visual understanding of the micro-relief

Available on website http://www.wrc.org.za 67 ISSN 0378-4738 (Print) = Water SA Vol. 36 No. 1 January 2010 ISSN 1816-7950 (On-line) = Water SA Vol. 36 No. 1 January 2010 JANAUER GA (1996) Macrophytes, hydrology, and aquatic ecotones: RAPER JF and KELK B (1991) Three-dimensional GIS. In: Maguire a GIS-supported ecological survey. Aquat. Bot. 58 379-391. DJ, Goodchild MF and Thind DV (eds.) Geographic Information JANSSEN R, GOOSEN H, VERHOEVEN ML, VERHOEVEN JTA, Systems: Principles and Applications. Longman: London. 1 OMTZIGT AQA and MALTBY E (2005) Decision support for 299-316. integrated wetland management. Environ. Modell. Software 20 SMITH N (2009) Personal communication. Department of Water 215-229. Affairs and Forestry, Potchefstroom, South Africa. JEWITT GPW, GARATT, JA, CALDER, IR and FULLER L (2004) SOULSBY C, TETZLAFF D, RODGERS P, DUNN S and WALDRON Water resources planning and modelling tools for the assessment S (2006) Runoff processes, stream water residence times and con- of land use change in the Luvuvhu Catchment, South Africa. Phys. trolling landscape characteristics in a meso-scale catchment: An Chem. Earth 29 1233-1241. initial evaluation. J. Hydrol. 325 197-221. KO C and CHENG Q (2004) GIS spatial modeling of river flow and SUI DZ and MAGGIO RC (1999) Integrating GIS with hydrological precipitation in the Oak Ridges Moraine area, Ontario. Comput. modelling: practices, problems, and prospects. Comput. Environ. Geosci. 30 379-389. Urban Syst. 23 33-51. LACROIX MP, MARTZ LW, KITE GW and GARBECHT J (2002) SUN G, RANSON KJ, KHARUK VI and KOVACS K (2003) Vali Using digital terrain analysis modeling techniques for the param- dation of surface height from shuttle radar topography mission eterization of a hydrologic model. Environ. Modell. Software 17 using shuttle radar altimeter. Remote Sensing Environ. 88 401-411. 127-136. TAYLOR F (2009) NASA/Japan Release Terrain Data – Google MINNAAR A (2009) Personal communication. Department of Water Earth’s is Better URL: http://www.gearthblog.com (Accessed 17 Affairs and Forestry, Sub Directorate: Remote Sensing, Pretoria. August 2009). PORTOGHESE I, URRICCHIO V and VURRO M (2005) A GIS tool WINDE F (2006) Challenges for sustainable water use in dolomitic for hydrogeological water balance evaluation on a regional scale in mining regions of South Africa – a case study of uranium pollu- semi-arid environments. Comput. Geosci. 23 15-27. tion. Part I: Sources and pathways. Phys. Geogr. 27 333-347.

68 Available on website http://www.wrc.org.za ISSN 0378-4738 (Print) = Water SA Vol. 36 No. 1 January 2010 ISSN 1816-7950 (On-line) = Water SA Vol. 36 No. 1 January 2010