Improved Remote Sensing Methods to Detect Northern Wild Rice (Zizania palustris L.)

Kristen O’Shea,Vanesa Martin,Timothy Mayer,Eli Simonson,Joshua Knopik,Jillian Laroe,Nicholas Young,Paul Evangelista,Daniel Carver,Paul Radomski,Colin K Khoury,Anthony Vorster,Anthony Kern

doi:10.3390/rs12183023

Abstract

Declining populations of Zizania palustris L. (northern wildrice, or wildrice) during the last century drives the demand for new and innovative techniques to support monitoring of this culturally and ecologically significant crop wild relative. We trained three wildrice detection models in R and Google Earth Engine using data from annual aquatic vegetation surveys in northern Minnesota. Three different training datasets, varying in the definition of wildrice presence, were combined with Landsat 8 Operational Land Imager (OLI) and Sentinel-1 C-band synthetic aperture radar (SAR) imagery to map wildrice in 2015 using random forests. Spectral predictors were derived from phenologically important time periods of emergence (June–July) and peak harvest (August–September). The range of the Vertical Vertical (VV) polarization between the two time periods was consistently the top predictor. Model outputs were evaluated using both point and area-based validation (polygon). While all models performed well in the point validation with percent correctly classified ranging from 83.8% to 91.1%, we found polygon validation necessary to comprehensively assess wildrice detection accuracy. Our practical approach highlights a variety of applications that can be applied to guide field excursions and estimate the extent of occurrence at landscape scales. Further testing and validation of the methods we present may support multiyear monitoring which is foundational for the preservation of wildrice for future generations.

Highlights

Northern wildrice (Zizania palustris L.), once abundant across the lakes of the Upper Midwest, has largely disappeared from the southern portions of its range during the last century [1,2]
Wildrice is fundamental to the habitat and diets of native waterfowl, fish [5], and 17 species of wildlife that are listed by the Minnesota Department of Natural Resources (MN DNR) as “species of greatest conservation need” [3]
The dominant taxa model predicted the largest area of wildrice presence and provided the most inclusive range, while having the highest false positive overlap rate out of the models (Figure 5)

Summary

Introduction

Northern wildrice (Zizania palustris L.), once abundant across the lakes of the Upper Midwest, has largely disappeared from the southern portions of its range during the last century [1,2]. Recognized as the state grain of Minnesota, conservation of this culturally, ecologically, and economically significant species is critical [3]. Indigenous peoples of the northern Great Lakes region have a strong cultural connection to wildrice and have harvested it for over two thousand years [4]. The annual harvest is still a sacred tradition to this day, which is critical for indigenous economic and food security [3]. Wildrice is fundamental to the habitat and diets of native waterfowl, fish [5], and 17 species of wildlife that are listed by the Minnesota Department of Natural Resources (MN DNR) as “species of greatest conservation need” [3]. Including hand-harvested wildrice, Minnesota’s annual yield ranges from four to eight million pounds [3]. Native stands of wildrice are considered resources with the potential to improve crop breeding amidst changes in climate and emerging pests and diseases [7]

Methods

Results

Conclusion