Development of profitable milk production systems for northern Australia: a field assessment of the productivity of five potential farming systems using farmlets.

Farmlets, each of 20 cows, were established to field test five milk production systems and provide a learning platform for farmers and researchers in a subtropical environment. The systems were developed through desktop modelling and industry consultation in response to the need for substantial increases in farm milk production following deregulation of the industry. Four of the systems were based on grazing and the continued use of existing farmland resource bases, whereas the fifth comprised a feedlot and associated forage base developed as a greenfield site. The field evaluation was conducted over 4 years under more adverse environmental conditions than anticipated with below average rainfall and restrictions on irrigation. For the grazed systems, mean annual milk yield per cow ranged from 6330 kg/year (1.9 cows/ha) for a herd based on rain-grown tropical pastures to 7617 kg/year (3.0 cows/ha) where animals were based on temperate and tropical irrigated forages. For the feedlot herd, production of 9460 kg/cow.year (4.3 cows/ha of forage base) was achieved. For all herds, the level of production achieved required annual inputs of concentrates of similar to 3 t DM/animal and purchased conserved fodder from 0.3 to 1.5 t DM/animal. This level of supplementary feeding made a major contribution to total farm nutrient inputs, contributing 50% or more of the nitrogen, phosphorus and potassium entering the farming system, and presents challenges to the management of manure and urine that results from the higher stocking rates enabled. Mean annual milk production for the five systems ranged from 88 to 105% of that predicted by the desktop modelling. This level of agreement for the grazed systems was achieved with minimal overall change in predicted feed inputs; however, the feedlot system required a substantial increase in inputs over those predicted. Reproductive performance for all systems was poorer than anticipated, particularly over the summer mating period. We conclude that the desktop model, developed as a rapid response to assist farmers modify their current farming systems, provided a reasonable prediction of inputs required and milk production. Further model development would need to consider more closely climate variability, the limitations summer temperatures place on reproductive success and the feed requirements of feedlot herds.

Automating Crop Damage Assessments with Unmanned Aerial Systems and Machine Learning

Agricultural pests are responsible for millions of dollars in crop losses and management costs every year. In order to implement optimal site-specific treatments and reduce control costs, new methods to accurately monitor and assess pest damage need to be investigated. In this paper we explore the combination of unmanned aerial vehicles (UAV), remote sensing and machine learning techniques as a promising methodology to address this challenge. The deployment of UAVs as a sensor platform is a rapidly growing field of study for biosecurity and precision agriculture applications. In this experiment, a data collection campaign is performed over a sorghum crop severely damaged by white grubs (Coleoptera: Scarabaeidae). The larvae of these scarab beetles feed on the roots of plants, which in turn impairs root exploration of the soil profile. In the field, crop health status could be classified according to three levels: bare soil where plants were decimated, transition zones of reduced plant density and healthy canopy areas. In this study, we describe the UAV platform deployed to collect high-resolution RGB imagery as well as the image processing pipeline implemented to create an orthoimage. An unsupervised machine learning approach is formulated in order to create a meaningful partition of the image into each of the crop levels. The aim of this approach is to simplify the image analysis step by minimizing user input requirements and avoiding the manual data labelling necessary in supervised learning approaches. The implemented algorithm is based on the K-means clustering algorithm. In order to control high-frequency components present in the feature space, a neighbourhood-oriented parameter is introduced by applying Gaussian convolution kernels prior to K-means clustering. The results show the algorithm delivers consistent decision boundaries that classify the field into three clusters, one for each crop health level as shown in Figure 1. The methodology presented in this paper represents a venue for further esearch towards automated crop damage assessments and biosecurity surveillance.