Wilfredo Robles

Potential for Remote Sensing to Detect and Predict Herbicide Injury on Waterhyacinth (Eichhornia crassipes)

Abstract 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

Estimating the Biomass of Waterhyacinth (Eichhornia crassipes) Using the Normalized Difference Vegetation Index Derived from Simulated Landsat 5 TM

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.

Exploiting Hyperspectral Hypertemporal Imagery with Feature Clustering for Invasive Species Detection

This paper presents a feature extraction method for exploiting hyperspectral hypertemporal data and applies the new method to the problem of invasive species detection. By definition, hyperspectral hypertemporal imagery is very high dimensional data, and dimensionality reduction will play a critical role in utilizing such data. We present a feature extraction method that takes advantage of the high correlation among elements of the spectral and temporal feature space. This high correlation can be attributed to the premise that the changes in the reflectance of closely spaced wavelengths do not always change dramatically over short periods of time. The proposed feature clustering method is based on the assumption that adjacent elements in the spectro-temporal feature space are highly correlated and can be grouped together to form lower- dimensional feature spaces. The proposed hyperspectral hypertemporal feature clustering method is tested and validated within an invasive vegetation detection application. The hypothesis is that as time progresses, the spectral response of different plant species change differently. Thus, there should be hyperspectral hypertemporal features that can be used to discriminate between the vegetative species. Additionally, the results of the feature clustering method can be used to determine which regions of the spectrum and which collection dates are optimum for the given invasives detection problem.