Angler apps as a source of recreational fisheries data: opportunities,challenges and proposed standards

Recreational fisheries surveys are limited in time and place in many countries. This lack of data limits scientific understanding and sustainable management. Smartphone applications (apps) allow anglers to record the details of their fishing trips and catches. In this study, we describe the opportunities and challenges associated with angler apps as a source of recreational fisheries data, and propose minimum standards for data collection via angler apps. Angler apps are a potentially valuable source of conventional and novel data that are both frequent and extensive, and an opportunity to engage anglers through data sharing and citizen science. Realizing this potential requires that we address significant challenges related to angler recruitment and retention, data quality and bias, and integration with existing fisheries programmes. We propose solutions to each of these challenges. Given that the angler app market is diverse, competitive and unpredictable, we emphasize minimum standards for data collection as a way to ensure large and reliable data sets that can be compared and integrated across apps. These standards relate to trips and catches, and angler demographics and behaviour, and should be supported through consultation and research. Angler apps have the potential to fundamentally change how anglers interact with the resource and with management.     

基于智能手机的棉花苗情调查与决策支持系统

针对新疆棉花生产过程中基层植保力量不足的问题,该文设计开发了棉花苗情调查与决策支持系统。该系统基于"云到端"模式,云端采用基于角色的访问控制(role-based access control,RBAC)解决多用户权限分配与菜单管理问题,提供数据管理分析、专家决策支持服务;用户端基于智能手机提供数据采集、专家决策意见接收功能。云到端之间的数据通信采用表述性状态转移(representational state transfer,REST)架构,提供RESTful的数据访问服务,解决系统之间松耦合、可扩展等问题。系统在新疆生产建设兵团第六师105团经过3 a的应用,结果表明:该调查系统能够快速获取文字和图片2种形式的棉田苗情信息,并能够方便的通过农田地图、柱状图、曲线图、Excel表格等多种方式对数据进行有效的分析和展示,大大提高了工作效率。通过搭起基层农户与生产管理者、农学专家之间的信息快速传输通道,加快了知识转化,减轻了基层植保压力。该系统加以扩展即可用于其他农作物的田间调查,同时这种调查模式稍加改造还可应用于林业、畜牧业等多个行业的信息普查。     

基于智能手机和人工神经网络反演小微水体水质参数研究

通过智能手机反演进行环境监测越来越受到关注，目前研究主要利用可见光反射率进行光学活性参数反演。本文基于水质现场监测数据，同步应用智能手机拍照，通过偏振镜、手机望远镜、不同规格滤波片、24色标准色卡，获取水体图像信息，结合逐步回归与人工神经网络方法，开展了长三角地区典型小微水体水质光学参数叶绿素和浊度以及非光学参数可溶性有机碳（dissolved organic carbon，DOC）反演监测。结果表明，研究区水体总体DOC浓度值变异范围在2.73-16.90 mg/L，浊度变异范围在6.53-91.10 NTU，叶绿素a浓度值变异范围在0.36-245.47 μg/L。通过逐步回归方法提取了水体DOC浓度的五个图像特征参数为R1＇、B/G2＇、R2"、R4"、B/G6＇，浊度图像特征参数为B/R3＇、G5"、R6"，叶绿素a的图像特征为B/G1＇、R2"、B/G4＇。结合人工神经网络模型反演水体水质参数，DOC浓度纳什系数NSE为0.62，浊度NSE为0.65，叶绿素NSE为0.67，具有较高的反演精度。本研究构建了基于智能手机反演水质光学参数的方法，并探讨了非光学参数反演的可行性，为后续开发APP应用程序和反演水质参数提供了基础和依据。