IMPACT OF DIGITAL TERRAIN ELEVATION DATA (DTED) RESOLUTION ON 
TERRAIN VISUALIZATION: SIMULATION VS. REALITY 
Louis A. Fatale 
James R. Ackeret 
U.S. Army Topographic Engineering Center 
ATTN: CETEC-CA-S, Fort Belvoir, VA, USA 22060-5546 
ABSTRACT: 
This study provides evidence of the influence of DTED resolution (in regions 
with different terrain roughness) on Army terrain data applications. The 
report documents results of the U.S. Army Topographic Engineering Center's 
(TEC) DTED resolution analysis including extensive field work and terrain 
visualization undertaken to investigate DTED integrity in comparison to the 
actual terrain. Finally, the report offers conclusions regarding DTED support 
for Army tactical and simulation applications. 
KEY WORDS: DTED, Resolution, Terrain, Visualization, Simulation, Sigma-t 
INTRODUCTION 
The following paper is a synopsis of a study 
that evolved from U.S. Army Topographic 
Engineering Center (TEC) special report number 
6, entitled " Digital Terrain Elevation Data 
(DTED) Resolution and Requirements Study: 
Interim Report" (Nov. 1990). A majority of 
users throughout the Mapping, Charting and 
Geodesy (MC&G) community have at one time or 
another expressed their concerns about the 
quality (resolution and accuracy) and quantity 
(worldwide coverage) of DTED. These concerns 
have focused on several issues such as cost of 
producing high resolution data, availability and 
acquisition of source material, and time 
required to produce a worldwide database. As 
DTED is employed in a majority of digital 
applications, it is the foundation product that 
most users require. Given this stature, an 
understanding of its limitations and appropriate 
applications is of utmost importance. 
PURPOSE /SCOPE 
This paper documents results of a TEC DTED 
resolution analysis for determination of the 
most desirable DTED resolution for various Army 
tactical and simulation applications. The 
terrain visualization portion of the study has 
been synopsized for this paper. 
METHODOLOGY 
The methodology employed in this effort was to 
select several geographic regions (Figure 1) 
with different terrain roughnesses. This 
evaluation attempts to show the comparison of 
these regions using simulated terrain for three 
DTED Levels: Level 1, Level 2, and the Level 2 
Downsampled (Level 2 (D)). The Level 2 (D) was 
produced by merely thinning the Level 2 to match 
the same post spacing as Level 1. Sigma-t 
values of DTED, which indicate the standard 
deviation of terrain height, were used to 
determine terrain roughness categories and 
respective geographic areas of interest. They 
are as follows: 
Region Terrain Roughness Sigma-t 
Classification (feet) 
Iran Very rough >800 
California Rough 200-800 
Maine Moderate 60-200 
Arkansas Smooth <60 
938 
The specific Sigma-t values for the above study 
areas are detailed in Table 1. 
This report includes a visual display analysis 
for each of the various terrain roughness types. 
Perspective views of these terrain types were 
generated on a Terrain Visualization Testbed 
system at USATEC for comparative analysis of 
DTED Levels 1 and 2. Field data, collected on 
site for varying terrain roughness 
classifications, were also incorporated into the 
investigation. 
DTED SOURCES 
Selected DTED coverage representing the range of 
terrain roughness classifications was obtained 
from the Defense Mapping Agency (DMA). All of 
the DTED in this analysis was compiled from 
photo source. Use of this common source allowed 
us to focus on compilation resolution, the key 
variable of interest. 
QUALITATIVE VISUAL ANALYSIS 
Hardcopy perspective area plots were 
produced for the following specific geographic 
areas: Qasr Od Dasht, Iran; Redding, 
California; Millinocket, Maine and El Dorado, 
Arkansas (Figure 1). Corresponding field data 
were collected and compared to the perspective 
plots as described below. 
Field Study 
In the interim report referred to earlier and in 
other similar analyses throughout the joint 
services, DTED resolution is compared in terms 
of overall definition of features and surface 
roughness for a number of applications. Because 
of its higher resolution, DTED Level 2 is 
usually assumed to be the "real world" model or 
as near to reality as possible. To enhance the 
perspective view analyses for this study, and to 
create a true "real world" control mechanism, a 
field team visited the three domestic study 
areas to record their surface conditions by 
photograph. These photographs were taken 
throughout each study area, so as to fully 
represent the terrain characteristics of that 
particular area. 1:24,000 scale USGS 
quadrangles were utilized to assure positioning 
and a compass was used to determine the azimuth 
of each photograph around the chosen site. For 
this synopsis, one site per study area is 
illustrated.