Robles W.,University of Puerto Rico at San Juan |
Robles W.,Geosystems Research Institute |
Madsen J.D.,University of California at Davis |
Wersal R.M.,Alpharetta Innovation and Technology Center
Invasive Plant Science and Management | Year: 2015
Waterhyacinth is a free-floating aquatic weed that is considered a nuisance worldwide. Excessive growth of waterhyacinth limits recreational use of water bodies as well as interferes with many ecological processes. Accurate estimates of biomass are useful to assess the effectiveness of control methods to manage this aquatic weed. While large water bodies require significant labor inputs with respect to ground-truth surveys, available technology like remote sensing could be capable of providing temporal and spatial information from a target area at a much reduced cost. Studies were conducted at Lakes Columbus and Aberdeen (Mississippi) during the growing seasons of 2005 and 2006 over established populations of waterhyacinth. The objective was to estimate biomass based on nondestructive methods using the normalized difference vegetation index (NDVI) derived from Landsat 5 TM simulated data. Biomass was collected monthly using a 0.10m2 quadrat at 25 randomly-located locations at each site. Morphometric plant parameters were also collected to enhance the use of NDVI for biomass estimation. Reflectance measurements using a hyperspectral sensor were taken every month at each site during biomass collection. These spectral signatures were then transformed into a Landsat 5 TM simulated data set using MatLab® software. A positive linear relationship (r2 = 0.28) was found between measured biomass of waterhyacinth and NDVI values from the simulated dataset. While this relationship appears weak, the addition of morphological parameters such as leaf area index (LAI) and leaf length enhanced the relationship yielding an r2 = 0.66. Empirically, NDVI saturates at high LAI, which may limit its use to estimate the biomass in very dense vegetation. Further studies using NDVI calculated from narrower spectral bands than those contained in Landsat 5 TM are recommended. Nomenclature: Waterhyacinth, Eichhornia crassipes (Mart.) Solms EICCR. Management Implications: Typically, the biomass of waterhyacinth is estimated using quadrats with a specific unit area placed over the plant mat. However, it is labor and time intensive to collect and process samples. Moreover, this method is destructive because it removes plant material from the system which affects long term studies of plant growth. The normalized difference vegetation index (NDVI) is a well-known vegetation index that can be used to monitor aquatic plants. However, limitations due canopy complexity during the growing season often limit its use. Based on the results, NDVI alone is not sufficient to estimate the biomass of waterhyacinth. The poor predictive performance of band 4, as well as canopy complexity related to waterhyacinth phenology during the growing season and vegetation cover/water background ratio likely affected the performance of NDVI. According to this study, measuring morphometric parameters such as leaf area index may enhance the performance of NDVI derived from Landsat 5 TM or other multispectral sensors with same spectral resolution. Therefore, the sole use of NDVI from Landsat 5 TM is not recommended to estimate the biomass of waterhyacinth. It is suggested that large-scale waterhyacinth management would consider NDVI derived from other multispectral sensors (e.g. Landsat 8 OLI). Current results could be useful to test new multispectral or hyperspectral sensors for aquatic vegetation management. © Weed Science Society of America.
Robles W.,University of Puerto Rico at Mayaguez |
Madsen J.D.,Geosystems Research Institute |
Wersal R.M.,Geosystems Research Institute
Invasive Plant Science and Management | Year: 2010
Many large-scale management programs directed toward the control of waterhyacinth rely on maintenance management with herbicides. Improving the implementation of these programs could be achieved through accurately detecting herbicide injury in order to evaluate efficacy. Mesocosm studies were conducted in the fall and summer of 2006 and 2007 at the R. R. Foil Plant Science Research Center, Mississippi State University, to detect and predict herbicide injury on waterhyacinth treated with four different rates of imazapyr and glyphosate. Herbicide rates corresponded to maximum recommended rates of 0.6 and 3.4 kg ae ha-1 (0.5 and 3 lb ac-1) for imazapyr and glyphosate, respectively, and three rates lower than recommended maximum. Injury was visually estimated using a phytotoxicity rating scale, and reflectance measurements were collected using a handheld hyperspectral sensor. Reflectance measurements were then transformed into a Landsat 5 Thematic Mapper (TM) simulated data set to obtain pixel values for each spectral band. Statistical analyses were performed to determine if a correlation existed between bands 1, 2, 3, 4, 5, and 7 and phytotoxicity ratings. Simulated data from Landsat 5 TM indicated that band 4 was the most useful band to detect and predict herbicide injury of waterhyacinth by glyphosate and imazapyr. The relationship was negative because pixel values of band 4 decreased when herbicide injury increased. At 2 wk after treatment, the relationship between band 4 and phytotoxicity was best (r2 of 0.75 and 0.90 for glyphosate and imazapyr, respectively), which served to predict herbicide injury in the following weeks. Nomenclature: Glyphosate; imazapyr; waterhyacinth, Eichhornia crassipes (Mart.) Solms EICCR © 2010 Weed Science Society of America.
Cox M.C.,Geosystems Research Institute |
Cox M.C.,Virginia Polytechnic Institute and State University |
Wersal R.M.,Geosystems Research Institute |
Wersal R.M.,Lonza Inc. |
And 3 more authors.
Invasive Plant Science and Management | Year: 2014
Alligatorweed, waterhyacinth, and hydrilla are three nonnative aquatic species of concern in the Ross Barnett Reservoir near Jackson, MS. Point-intercept surveys were conducted on the reservoir from 2005 to 2010 to monitor native and nonnative species' distributions and assess herbicide treatment efficacy across the reservoir. Foliar applications of 2,4-D, glyphosate, imazapyr, and diquat were made during summer months for emergent and free-floating vegetation, whereas submersed applications of liquid copper and granular fluridone were applied in spring and late summer for subsurface hydrilla populations. American lotus is the native species that has been observed the most throughout the survey years, with occurrence frequencies averaging between 17 and 27%. Alligatorweed populations significantly decreased from 21% in 2005 to 4% in 2006; however, they consistently increased in the next 4 yr to 12% occurrence in 2010. Waterhyacinth occurrence has remained relatively constant over the study period, averaging below 10% occurrence. Hydrilla was discovered in the reservoir in late 2005 and has remained below 2% in frequency of occurrence since 2006. Suppression of these nonnative species has been attributed to rigorous monitoring and herbicide applications conducted on the reservoir since 2005. A logistic regression model indicated that as native species richness increased, the likelihood of a nonnative species occurring also increased. Nomenclature: Alligatorweed, Alternanthera philoxeroides (Mart.) Griseb., American lotus, Nelumbo lutea Willd., hydrilla, Hydrilla verticillata (L. f.) Royle, waterhyacinth, Eichhornia crassipes (Mart.) Solms.
Wersal R.M.,Geosystems Research Institute |
Madsen J.D.,Geosystems Research Institute
Invasive Plant Science and Management | Year: 2010
Parrotfeather is an invasive, aquatic plant in the United States that is native to South America. It has impaired the use of water bodies throughout the United States and is difficult to control, despite using a variety of management techniques. Our objectives were to examine the efficacy of subsurface applications of seven herbicides labeled for aquatic use and to compare those applications to herbicides that can also be applied to emergent foliage. A replicated mesocosm study was conducted in 378-L (100-gal) tanks beginning in August 2007 and repeated during the same period in 2008. The maximum and half-maximum labeled rates of copper chelate, diquat, endothall, fluridone, triclopyr, and carfentrazone-ethyl were applied to the water column in designated mesocosms. The maximum labeled rate for foliar applications of diquat, triclopyr, and 2,4-D were used to compare treatment methods. Six weeks after treatment (WAT), copper, endothall, fluridone, and carfentrazone-ethyl did not achieve 90% control; in fact, control was less than 50% for each herbicide, and therefore, the herbicides were not considered efficacious for controlling parrotfeather. Diquat at all rates and application methods resulted in 70 to 90% biomass reduction. Triclopyr, with both the highest aqueous concentration and foliar application, resulted in an 84 and 86%, respectively, reduction in biomass at 6 WAT. The foliar application of 2,4-D was the only herbicide and application method that resulted in ≥90% biomass reduction of parrotfeather. In these studies, regrowth occurred in all tanks regardless of herbicide or treatment method, indicating multiple applications would be necessary to provide longer-term plant control. Future research should identify possible herbicide combinations or timing of applications to maximize treatment efficacy. © Weed Science Society of America 2010.
Alarcon V.J.,Geosystems Research Institute |
O'Hara C.G.,Geosystems Research Institute
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) | Year: 2010
This paper investigates the effect of land use and digital elevation models spatial resolution and scale on the simulation of stream flow in two coastal watersheds located in the Mississippi Gulf Coast (USA). Four elevation datasets were used: USGS DEM, NED, NASA's SRTM and IFSAR (300, 30, 30, and 5 meter resolution, respectively). Three land use datasets were included in this study: USGS GIRAS, NLCD, and NASA MODIS MOD12Q1 (400, 30, and 1000 m resolution, correspondingly). The Hydrological Program Fortran (HSPF) was used for estimating stream flow in the two watersheds. Results showed that swapping datasets in a factorial design experiment produce equivalent statistical fit of measured and simulated stream flow data. The results also showed that HSPF-estimated stream flows are not sensitive to scale and spatial resolution of the datasets included in the study. © 2010 Springer-Verlag Berlin Heidelberg.