Amots Hetzroni
| تَنظِيم | ARO - Volcani center |
|---|---|
| Organization type | Research Institution |
| الدولة | Israel |
This member participated in the following Forums
المنتدى Forum: "Building the CIARD Framework for Data and Information Sharing" April, 2011
Question 1: What are we sharing and what needs to be shared?
To my approach, agricultural data should managed by the data owner, aggregated in local/regional data centers. Thus, unlike peer reviewed published articles, and because we are looking into ample amount of data, we should expect data validation and veracity to vary from one source to another.
I think that we should think of some mechanisms that will rank data for completeness, integrity, etc. This will allow us to reach more information, and if the source is marked as fickle, yet appealing by its content, one might approach directly the data manager for clarifications.
The opportunity to access data from various resources opens new doors for meta-analysis studies. In a study aiming to model for soil-borne disease, we faced lack of data trying to verify our multi-layer prediction. We were able to track published data from two studies done in different locations on the globe and the contribution was valuable. I can only imagine the possibilities to compare and test models on ample data.
I argue that we have two sides and both should gain from the information. Assuming that the majority of the prospective players are not researchers but rather users who are looking for an answer to a problem, they do not have the time and resources to learn the genre of a new system composed by a remote foreign society. The knowledge should be presented in a way that is easy for them to grasp.
The clients for information diversify geographically, economically, socially, from a wide range of literacy, and practice and there are language barriers amongst. Therefore, a unified information system cannot provide an answer to all players. Almost none would be interested to know that the north-east edge of a certain tomato field was infested with Helicoverpa armigera by the end of April. But for the neighbor grower, this information might be indispensable. If integrated pest management is practiced in the region, then the moderators would want to know what pesticide was applied. Crop growers fifty km away would be alerted that there is a changed trend in pest population and might start with prophylactic measures. The target market might be interested to know that crop reduction might be expected and would look for alternative sources.
Everyone in the food chain uses as much information as they can get and digest. The level of details should harmonize with the needs and abilities. The data that is available to the user should match his language, terminology and cultural associations.
A network of distributed information systems allows local centers to manage their information efficiently. They can efficiently ensure data veracity and integrity. These local centers can be located and managed in farms or regional centers. They should be able to provide information services to queries from other centers, and that requires to have adequate thesaurus systems to ensure that differences in terminology would not render the erroneous information.
The fear from sharing information can better be managed locally where culture issues can be handled by natives to the region, yet this is certainly a long process.
Hello all,
My name is Amots Hetzroni and I work in the institute of agricultural engineering which is part of a government research organization (ARO) in Israel. My major interest is the use of ICT, specifically data management in agriculture.
Recognizing that knowledge and information are key-factors in pest management decision making, we put an effort into establishing a collection, storage and dissemination system for pest scouting data, as part of a decision support system for pest management. We started by developing a data collection system based on pocket-PC’s. The concept was proven to be feasible; and has thus been accepted and used by the pest scouts. Yet, it lacked two-way communication to provide the feedback required in order to support decision making in the field. Therefore, the units were replaced by cellular phones linked to the repository data server.
The interface is rather friendly, and we continually working on minimizing the load on the user. Upon login, local data tables are being updated from the main server, data such as a list of plots and crops pertinent to specific user. The scouting records, including time and location are transferred, upon communication availability, to the server. The repository was designed for internet interface to respond to spatial queries. And interfaces are being developed. Predefined queries and filters, such as history of infestations, are available for the end user from the cellular terminal.
We make efforts to add more players who will use the system: farmers, extension officers, decision making in the region, researchers and more. And we make slow progress.
Some of the difficulties include the maintenance of valid data such as crop schedule and pesticide application, getting some farmers to contribute their data, training players to make use of the system, and more.
We have also created a regional repository for activities in date palm farms, where we linked together the activities and dates’ quality. We intend to use the database as tool to support precision horticulture and for future data mining when searching for quality issues that are associated with the activities, methods and timing. At the moment, the farmers utilize mainly the detailed data on machinery usage and manual labor for economic analysis.