Search USGS⌕ Search

Geology topics

Robert H. Haas

Publications and source records attributed to Robert H. Haas.

9 recordsLinked to original sources
USGS1992

Landsat thematic mapper products for rangeland assessment

Landsat Thematic Mapper (TM) data are useful for mapping rangeland resources worldwide. Spatial resolution and spectral characteristics of the data are adequate for performing a variety of range‐related tasks, including monitoring ecological change. The unique spectral characteristics of TM data are important in the development of products for manual interpretation of images to be used in rangeland assessment. Procedures were developed for using a histogram‐based, data‐dependent stretch of multispectml data values for producing site‐specific TM image products. The process optimizes image information for a given data set without a priori knowledge of the imaged site. Manual interpretation of TM products for rangeland assessment has many applications and will be aided by a standardized procedure for compositing spectral bands.

Geocarto International↗
USGS1992

Analyzing vegetation dynamics of land systems with satellite data

Large area assessment of vegetation conditions is a major requirement for understanding the impact of weather on food, fiber, and forage production. The distribution of vegetation is largely associated with climate, terrain characteristics, and human activity. The interpretation of vegetation dynamics from satellite data can be improved by stratifying the land surface into ecoregions. The Soil Conservation Service, U.S. Department of Agriculture, has developed a system for mapping major land resource areas (MLRA) that groups land areas in the United States on the basis of climate, physiography, land use, and land cover characteristics. In 1989, the U.S. Geological Survey used National Oceanic and Atmospheric Administration weather satellite data to conduct a biweekly assessment of vegetation conditions in 17 western states. Advanced Very High Resolution Radiometer data were acquired daily, and were geographically registered, and the normalized difference vegetation index (NDVI) was computed for the Western United States during the 1989 growing season. Fifteen biweekly NDVI data sets were used to evaluate MLRA's as an appropriate stratification for monitoring and interpreting vegetation conditions in the study area. The results demonstrate the feasibility of using MLRA's to stratify areas for monitoring phenological development and vegetation condition assessment within the growing season. Assessments of the NDVI at biweekly intervals are adequate for monitoring seasonal growth patterns on MLRA's where rangelands, forests, or cultivated agriculture are the primary resource type. Descriptive statistics are indicators of the uniformity or diversity of land use and land cover within an MLRA. Growing season profiles of the NDVI are characterized by the seasonal effects of climate on various land use and land cover classes.

Geocarto International↗
USGS1986

A new view for resource managers

In decades past, the rancher depended upon reports from cowboys to gather information he needed to make management decisions. Today, the vast open ranges of the cowboy era are mostly gone in the United States-fenced into pastures, paddocks, or fields that are now discrete management units. But fencing in the rangeland, while it has replaced much of the need for cowboys, has not replaced the need for information about the health and vigor of the forage on each parcel of land. Can a satellite, orbiting at more than 400 miles in space, serve this purpose? As ranchers and resource specialists are asked to make more and more complex management decisions, with less manpower for conducting inventories, they are wise to seek help in today's rapidly developing technologies. For the past few decades the range technician has accomplished most of his range assessment from a pickup truck, traveling periodically to each unit to determine its status. Now, satellite images of the Earth's resources might be able to help the modern range person do an even more efficient job of monitoring the availability of feed for livestock and wildlife. Yet some important questions need to be answered first. Can this new information source be used to evaluate the ecological condition of these lands? Or are satellite images of our Earth and its variety of landscapes just "pretty pictures," with little practical utility?

Rangelands↗
USGS1985

Change detection in rangeland environments using Landsat MSS data: A quantitative approach

A range forage utilization study on the Crow Creek Indian Reservation in central South Dakota provided the opportunity to use Landsat multispectral scanner (MSS) data for examining range condition trends. A procedure was developed to compare change in spectral reflectance over time for polygon areas, defined by resource type within management units. A t-test was used to evaluate changes in brightness and greenness within pastures between September 27, 1978, and September 18, 1983. The first principal component transformation from four-band MSS images for both dates was used as a measure of brightness. Greenness was measure using the second principal component transformation for both dates. Examination of the brightness date showed that the assumptions required for a valid t-test were met. The greenness data violated the assumption of independence between dates and was not used for trend comparisons. The t-values calculated from each polygon were coded into three groups: (1) those indicating significant brightness decrease, (2) those indicating significant brightness increase, and (3) those indicating no significant brightness change. Significance was determine at the 5-percent level. These results were formatted into an image, which is a preliminary product for evaluating range condition trends over a 5-year period.

South Dakota↗
USGS1982

Changes in vegetation and grazing capacity following honey mesquite control

Honey mesquite kill and suppression, vegetation response, and changes in grazing use and capacity were evaluated following brush control in north-central Texas. Tree grubbing was most effective for eliminating honey mesquite, but because of soil and plant damage the treatment did not increase grazing capacity or improve range condition compared to nontreated rangeland. Aerial application of 2,4,5-T + picloram was more effective in klllmg and defoliating honey mesquite than 2,4,5-T alone, but both treatments significantly increased forage production. The 2,4,5-T + picloram and 2,4,5-T sprays provided a 7 to 16% increase in grazing capacity over a 4-year period on light and heavy honey mesquite infested pastures, respectively.

Texas↗
USGS1982

Assessing mesquite-grass vegetation condition from Landsat

Landsat multispectral scanner (MSS) band values, band ratios, and vegetation index models were compared with selected rangeland vegetation parameters collected at six test sites within the honey mesquitellotebushlmixed grass association in north-central Texas. The comparisons at four dates showed that two vegetation index models, TV16 and GVI, are highly correlated (P = 0.01) with green yield, green cover, and plant moisture content. The green vegetation index (GVZ) developed by Kauth and Thomas (1976), was highly correlated and superior to other models in relationship to wet green yield, dry green yield, and cured vegetation cover. TV16, developed by Rouse et al. (1974), was more highly correlated with green vegetation cover and vegetation moisture content. Both TV16 and GVI are superior to other models in their relationship with green cover. None of the Landsat MSS parameters tested was significantly correlated with dry total yield, percent bare ground, or moisture of the soil measured at the surface or at a 20 cm depth. I t is concluded that Landsat MSS data are sensitive to seasonal changes in vegetation growth conditions and inherent ecological differences within a relatively unqorm vegetationlsoil system.

Texas↗
USGS1981

Evaluation of Landsat Multispectral Scanner data for mapping vegetated soil landscapes

Landsat multispectral scanner data for Brazos County, Texas, were evaluated in terms of effectiveness for classifying soils on vegetated landscapes at three times during the year: a time of normally adequate soil water, a time of expected soil water deficit, and a time when soil water is normally being replenished. Six test sites were used to evaluate LARSYS supervised and unsupervised classification of vegetated soil landscapes. Open grassland soils were best separated in the fall during a period when soil moisture was being replenished after the summer period of soil water deficit. Woodland soils were separated by Landsat data in late spring when adequate moisture was available. However, a high degree of accuracy was not achieved using Landsat for separating soil map units. Accurate separation of soil mapping units on vegetated landscapes was not possible during late summer when soil water was deficient. Selected soil properties important to plant growth were separable on the test sites using June and October Landsat data. Particle size and soil moisture regime were separated at both dates. Soils with argillic horizons were separated from soils without argillic horizons.

Texas↗