Get reliable rangeland science

In the state of Chihuahua, Mexico, new records of the invasion of Natal grass (Melinis repens) have been detected. This expansion is affecting livestock production as well as several ecosystems services, such as biodiversity. The objective of this study was to develop a prediction model to simulate the Natal grass distribution. The model takes into account the environmental factors where this grass is established. A total of 169 location points of Natal grass and 27 environmental variables were used to calibrate and validate the model. The software MaxEnt, version 3.3.3k and the technique of Principal Components Analysis (PCA) were used. Nineteen climatic variables were retrieved from the Wordclim database and 8 physical variables from each location point were obtained from thematic maps of the state. The model was evaluated throughout the ROC curve under the AUC = 0.949. The PCA selected 12 climatic variables. Precipitation of the season was the variable with the highest contribution for the model. The climate of locations with high probability of occurrence for Natal grass is a temperate semi-dry. No preference for soil texture was detected for the species. Most of the presence of Natal grass was located in places with air temperatures between 30 and 34 �C. The potential distribution resulted towards the short grassland prairie dominated by�Bouteloua gracilis.�The actual distribution of Natal grass covers 161,322 ha and will potentially increase to 458,000 ha. The developed model could be used as a tool for the elaboration of control or management programs for Natal grass.