Editor's Note: Sensors, as the "nerve endings of the information age," have penetrated every critical field of the socio-economic landscape. Since October 2025, China Electronics News has invited Guo Yuansheng, Deputy Director of the Science and Technology Committee of the Jiusan Society and Executive Vice Chairman of the China Sensor and IoT Industry Alliance, to open a column titled "Guo Yuansheng Explains Sensors". Focusing on eight major fields and scenarios—electric power, major equipment, intelligent manufacturing, smart agriculture, smart healthcare and big health, smart home appliances and consumer electronics, urban security, and low-altitude economy—the column has published articles such as "Stationed" on the Power Generation Side: Sensors Become the Cornerstone of Stable Operation of New Power Systems and Energy Storage Sensors Clarify Three Core Development Directions, receiving widespread attention and high praise from readers. This issue publishes the third article in the smart agriculture field, focusing on forestry and grassland monitoring sensors, elaborating on their application prospects, industrial status, and future suggestions to form industrial consensus and promote industrial development.
The global forestry and grassland industry is transforming from single ecological protection to the integrated development of "ecology + safety + industry." As the core component of the "space-air-ground" integrated monitoring network, intelligent sensors have become the key technology supporting refined management across the entire forestry and grassland domain. Currently, China is comprehensively advancing the strategic planning goals for deepening the smart forestry and grassland industry, explicitly setting the intelligent construction of the three major tasks—"ecological monitoring, safety prevention and control, and industrial efficiency enhancement"—as the innovation goals for high-quality development during the "15th Five-Year Plan". Combined with high-tech such as intelligent perception and AI (Artificial Intelligence), it aims to achieve comprehensive monitoring coverage, precise data feedback, rapid emergency response, and scientific disposal plans, adapting to higher requirements for future digitalization and intelligence across the entire forestry and grassland domain.
Traditional manual patrol or single-sensor monitoring models suffer from prominent issues such as limited coverage, delayed response, and fragmented data, making it difficult to adapt to the complexity of the forestry and grassland ecosystem. These complexities include monitoring parameter differences caused by tree species diversity and the randomness of wildlife activities; the urgency of safety prevention and control, such as real-time early warning needs for forest fire prevention and rapid diffusion prevention of pests and diseases; and the practicality of industrial development, such as precise management of understory economic crops and dynamic regulation of grassland carrying capacity. Against this backdrop, systematically sorting out sensor application solutions from ecological protection, safety prevention and control to the entire understory economy scenario, and clarifying the core parameters, technical basis, and application models under each scenario, plays a key role in promoting the intelligent upgrade of the forestry and grassland industry, building a solid ecological security barrier, and empowering the rural revitalization strategy.
I. Core Parameter System for Forestry and Grassland Monitoring
Building a scientific and comprehensive forestry and grassland monitoring parameter system is the prerequisite and foundation for achieving precise monitoring. Based on the development logic of the forestry and grassland industry—"safety first, ecology-based, and industry-empowered"—the monitoring parameter system can be divided into two major categories: general parameters for all forest types and special parameters for different forestry and grassland types. This ensures both the comprehensiveness of full-domain coverage and the targeted nature of specific scenarios.
General Monitoring Parameters for All Forest Types
General monitoring parameters for all forest types focus on the common needs of the forestry and grassland ecosystem. They are prioritized in the order of "safety prevention and control - ecological protection - environmental support" to ensure that core demands are implemented first. The core parameters for safety prevention and control are directly related to the survival bottom line of forestry and grassland resources and are the top priority of the monitoring system; basic parameters for ecological protection support the health assessment of the ecosystem; environmental support parameters provide scenario backgrounds for the first two categories, forming an organic whole.
In terms of core parameters for safety prevention and control, forest fire prevention and pest/disease early warning are the two core directions. Forest fire prevention parameters need to accurately capture early signals of fire occurrence. Referring to national standards such as GB/T 36743-2018 Forest Fire Danger Weather Grade, the surface temperature monitoring accuracy needs to reach ±1℃, and when the monitored value exceeds 60℃, a high-temperature warning is triggered; smoke concentration monitoring follows LY/T 2586-2016 Technical Specification for Forest Smoke Monitoring, using PM2.5 concentration ≥150μg/m³ as the smoke alarm threshold; open flame radiation intensity is based on the technical specifications for forest fire danger warning systems, and when the radiation power ≥10kW/m², it is judged as an open flame, triggering an emergency response. Pest and disease early warning parameters focus on environmental triggers and population density where pests and diseases are highly prevalent. Referring to LY/T 3156-2019 Technical Regulations for Forestry Pest and Disease Monitoring, when forest humidity ≥80% and the temperature is in the 25-30℃ range, it is a high-incidence environment for pests and diseases, requiring increased monitoring frequency; insect population density monitoring follows LY/T 2282-2014 Guidelines for Monitoring and Surveying Forestry Harmful Organisms, and when the number of insects per hundred trees ≥50, a prevention and control warning is triggered.
Basic parameters for ecological protection focus on two core dimensions: vegetation vitality and biodiversity. Among vegetation vitality parameters, the NDVI (Normalized Difference Vegetation Index) is the core indicator for assessing vegetation coverage and health status. According to HJ 192-2022 Technical Specification for Evaluation of Ecological Environment Status, an NDVI in the range of 0.2-0.8 corresponds to a healthy growth state of vegetation; below 0.2 indicates vegetation degradation, and above 0.8 requires attention to the fire risk caused by overly dense vegetation; leaf water content monitoring follows LY/T 3263-2021 Technical Specification for Monitoring Water Content of Forestry Vegetation, and when the leaf water content <50%, it is judged as drought stress, requiring the initiation of irrigation intervention. Biodiversity parameters focus on wildlife activity monitoring, using infrared camera monitoring technology. Referring to LY/T 2408-2015 Technical Regulations for Infrared Camera Monitoring of Wildlife, when the shooting frequency of a certain species is <3 times per month, attention should be paid to changes in its population size, and potential risks such as habitat destruction should be investigated.
Environmental support parameters provide basic environmental background data for the forestry and grassland ecosystem, mainly including rainfall and soil temperature and humidity. Rainfall monitoring follows GB/T 21978.2-2014 Precipitation Observation Specification Part 2: Tipping Bucket Rain Gauge, providing data support for water resource regulation and fire risk assessment; soil temperature and humidity monitoring refers to GB/T 35221-2017 Technical Specification for Soil Moisture Monitoring, accurately grasping soil moisture conditions and providing an environmental basis for vegetation growth and understory crop planting.
Special Monitoring Parameters for Different Forestry and Grassland Types
Different forestry and grassland types have significant differences in ecological characteristics and industrial functions, requiring targeted setting of special monitoring parameters to ensure the adaptability of monitoring solutions. Combined with the distribution characteristics of China's forestry and grassland resources, the focus is on covering five core scenarios: artificial forests, natural forests, grasslands, nature reserves, and the understory economy.
Artificial forests (including economic forests) take yield improvement and quality control as core goals, with special monitoring parameters focusing on growth and timber formation indicators. DBH (Diameter at Breast Height) monitoring follows GB/T 26424-2010 Measurement Methods for Tree Diameter at Breast Height, with a measurement accuracy of ±0.1cm to accurately grasp the tree growth rate; tree height monitoring uses LiDAR technology, referring to LY/T 1698-2007 Technical Regulations for Remote Sensing Survey of Forest Resources, to achieve rapid and accurate measurement; flowering and fruiting rate is the core output indicator for economic forests. According to NY/T 2312-2013 Technical Specification for Monitoring Flowering and Fruiting Rate of Economic Forests, when the flowering and fruiting rate ≥60%, it is judged as a high-yield standard, and if it is below this standard, issues such as pollination, water, and fertilizer need to be investigated.
Natural forests (including ancient and famous trees) focus on ecological protection and structural stability, with special monitoring parameters focusing on ecological structure and health status. Canopy density is the core indicator reflecting the ecological structure of natural forests. Following LY/T 1843-2009 Technical Regulations for Determination of Forest Canopy Density, the stability of the natural forest ecosystem is assessed by monitoring changes in canopy density; the monitoring of ancient and famous trees focuses on the degree of decay, using stress wave detection technology. Referring to LY/T 3213-2020 Technical Specification for Health Monitoring of Ancient and Famous Trees, when the decay rate >30%, reinforcement and protection measures need to be initiated to avoid the ancient trees falling and being damaged.
Grasslands take ecological restoration and carrying capacity balance as core goals, with special monitoring parameters focusing on degradation status and carrying capacity. Forage height monitoring follows NY/T 635-2022 Technical Specification for Grassland Monitoring, with a measurement accuracy of ±1cm. When the forage height <10cm, it indicates grassland degradation, and grazing ban or rest measures need to be implemented; the proportion of desertified area is a key indicator for assessing grassland ecological degradation. According to HJ 633-2012 Technical Specification for Evaluation of Ecological Environment Status, when the proportion of desertified area >20%, a desertification control project needs to be initiated.
Nature reserves focus on biodiversity protection and human interference prevention and control, with special monitoring parameters focusing on flagship species protection and human activity management. Among flagship species protection parameters, GPS collar trajectory monitoring technology is used, with a positioning accuracy of ≤5m, to real-time grasp the species' activity range and migration routes; taking the Giant Panda National Park as an example, the distribution density of staple bamboo also needs to be monitored. According to the monitoring specifications of the Giant Panda National Park, it ensures that the staple bamboo resources can meet the survival needs of the population. Human interference parameters focus on monitoring illegal logging and unauthorized entry. Illegal logging acoustic monitoring follows LY/T 3243-2020 Technical Specification for Acoustic Monitoring of Illegal Forestry Operations. Infrared triggered image monitoring requires a resolution of ≥2 megapixels to achieve precise capture and traceability of human interference behaviors.
As an important direction for enhancing the efficiency of the forestry and grassland industry, the understory economy has special monitoring parameters focusing on crop growth and the adaptability of the breeding environment. Among understory crop growth parameters, mushroom planting requires monitoring the temperature and humidity in the shed. Referring to NY/T 5333-2020 Technical Specification for Environmental Control of Edible Fungi Cultivation, a temperature of 18-22℃ and humidity of 85-90% is the optimal growth environment; Chinese medicinal material planting focuses on monitoring soil fertility. According to NY/T 1121.7-2014 Soil Testing Part 7: Determination of Soil Available Phosphorus, when the soil nitrogen content ≥1.0g/kg, it can meet the growth needs of most Chinese medicinal materials. Among breeding environment parameters, understory chicken farms need to monitor ammonia concentration, following GB/T 36195-2018 Environmental Quality Standard for Livestock and Poultry Breeding, with an ammonia concentration of ≤15mg/m³ as the qualified standard; beehive temperature monitoring refers to NY/T 3146-2017 Technical Specification for Bee Breeding Management, with 32-35℃ being the optimal breeding temperature for bees, ensuring the health of the bee colony and honey production quality.
II. Sensor Technology System for Forestry and Grassland Monitoring
Based on the core parameter requirements for forestry and grassland monitoring, a "space-air-ground" integrated sensor networking architecture is constructed. Through hierarchical sensor selection and precise deployment, full-scenario monitoring coverage for ecological protection, safety prevention and control, and the understory economy is achieved.
Space-Air-Ground Integrated Sensor Networking Architecture
Space-based satellites (Gaofen-5) transmit macroscopic data to regional gateways via Beidou short messages → Air-based UAVs (4G/5G) upload mesoscopic data in real time → Ground-based sensors network via LoRaWAN (transmission distance 5km), finally converging at the forestry and grassland big data platform, with an overall transmission delay of ≤30s.
1. Space-Based Sensor Layer: Focusing on Full-Domain Macroscopic Monitoring to Support Strategic Decision-Making
Hyperspectral satellite sensors (such as the visible shortwave infrared hyperspectral camera carried by Gaofen-5 02): spectral resolution ≤5nm, spatial resolution 30m, swath width 60km, capable of achieving vegetation type identification (accuracy ≥92%) and leaf area index inversion (accuracy ±0.3), with a data update cycle of 5 days, providing large-scale data support for forestry and grassland resource censuses.
Thermal infrared satellite sensors (such as the infrared payload of Fengyun-3G): temperature sensitivity ±0.2℃, spatial resolution 100m, scanning swath width 2000km, capable of real-time capture of abnormal surface temperatures in forest areas (temperature gradient >5℃/km), providing early warning of potential fire points 48 hours in advance, covering blind spots of manual patrol such as uninhabited areas and remote mountainous regions.
2. Air-Based Sensor Layer: Focusing on Regional Mesoscopic Monitoring to Bridge Macroscopic and Microscopic Levels
Airborne LiDAR sensors (Model: RIEGL VQ-1560i): point cloud density 100 points/㎡, ranging accuracy ±2cm, scanning angle 360°, capable of accurately measuring tree height (accuracy ±5cm) and canopy density (resolution 0.01), suitable for natural forest resource surveys and 3D modeling of ancient and famous trees. When the flight height is 300m, the coverage area per flight is ≥5km².
Multispectral UAV sensors (Model: MicaSense RedEdge-MX): includes 5 bands of blue, green, red, red edge, and near-infrared, spectral resolution 10nm, dynamic range 12-bit, capable of capturing changes in vegetation reflectance (such as a decrease in red edge band reflectance ≥15% caused by pine wood nematode disease), achieving early warning of pests and diseases. When the UAV flight height is 300m, the flight route planning adopts a "parallel route + 30% overlap rate" design, covering 1000 mu per flight, and a 2-hour endurance can complete 5 monitoring units.
3. Ground-Based Sensor Layer: Focusing on Point Microscopic Monitoring to Ensure Real-Time Control
Ground-based sensors follow the principle of "grid deployment + encryption in key areas," with a 500m×500m grid in mountainous areas and a 1000m×1000m grid in plains.
Fixed monitoring station integrated sensor groups: core configurations include meteorological sensors (Model: Campbell CS215, air temperature accuracy ±0.2℃, relative humidity accuracy ±2%RH), soil sensors (Model: Decagon 5TM, soil moisture accuracy ±2%vol, soil temperature accuracy ±0.5℃), and ecological sensors (such as infrared animal counters, detection distance 5m, recognition accuracy ≥85%). Data collection intervals are 1-5 minutes, transmitted via 4G/Beidou dual-mode, with a transmission delay of ≤10s, supporting 7×24 hours of uninterrupted monitoring.
Portable handheld sensors: such as infrared thermometers (Model: Fluke 62MAX+, temperature measurement range -20℃~600℃, accuracy ±1℃) and insect condition collection boxes (built-in macro camera, resolution 20 megapixels, supporting AI recognition of insect body features), suitable for on-site verification, enabling precise verification of abnormal areas (such as suspected fire points, concentrated areas of pests and diseases), making up for the detail blind spots of remote monitoring.
Sensor Selection and Deployment Solutions for Different Scenarios
1. Forest Fire Prevention Scenario: Adopting a Multi-Source Composite Early Warning Scheme of "Pyroelectric Infrared + Laser Smoke + AI Vision"
Pyroelectric infrared captures temperature anomalies → laser smoke sensors verify smoke concentration → AI vision distinguishes between fire smoke and cooking smoke through morphological analysis. This three-level verification logic can reduce the false alarm rate to ≤3%.
Core sensor configuration: Pyroelectric infrared array sensor (Model: FLIR Boson320): detection distance 150m, temperature resolution 0.05℃, frame rate 30Hz, dynamic range 14bit, output protocol SPI/I²C, protection level IP67, anti-electromagnetic interference capability (compliant with IEC61000-4-3 standard), capable of capturing surface temperature changes of 0.1℃, triggering high-temperature warnings (threshold ≥60℃); laser scattering smoke sensor: zero drift ≤±2μg/m³/year, operating temperature -20℃~60℃, anti-humidity interference (0-95% RH without condensation), PM2.5 measurement range 0-1000μg/m³, accuracy ±10μg/m³, response time ≤30s, alarm triggered when smoke concentration ≥150μg/m³; AI smart camera (Model: Hikvision DS-2TD2617B-6/V1): integrates thermal imaging and visible light dual modules, thermal imaging resolution 384×288, temperature sensitivity ±0.05℃, visible light resolution 2 megapixels, supports intelligent recognition of open flames/smoke (accuracy ≥98%), capable of distinguishing cooking smoke from fire smoke.
Deployment logic: In mountainous areas, deploy in a 500m×500m grid. Each monitoring point integrates a solar power supply system (battery capacity 20Ah, endurance ≥7 days of rainy days), with a bracket height of 3-5m to avoid vegetation occlusion; in plain forest areas, deploy 1 unit every 1000m along roads and forest edge lines, using buried cables + dual solar power supply, equipped with 4G/Beidou dual-mode transmission modules (data upload delay ≤10s) to form a linear early warning defense line; in key areas (such as around temples and forest area homestays), deploy with increased density, reducing the spacing to 300m, and simultaneously link with sound and light alarms (alarm decibel ≥110dB) to remind surrounding personnel to evacuate in time.
2. Pest and Disease Prevention and Control Scenario: Adopting a Collaborative Monitoring Scheme of "Intelligent Insect Trapping + Environmental Monitoring + Spectral Analysis"
Core sensor configuration: Intelligent insect condition forecasting lamp (Model: Top Cloud-agri TPCB-II-C6.0): insect trapping wavelengths of 365nm/400nm dual bands, insect trapping distance ≥20m, built-in insect body photography module (resolution 5 megapixels), uses YOLOv8 object detection algorithm, supports AI recognition of insect species (accuracy ≥90%), recognition accuracy for Monochamus alternatus ≥95%, supports offline recognition (local model storage capacity ≥16GB), can count insect population density (unit: heads/day); temperature, humidity, and spore capture sensors (Model: AWS620-L): temperature accuracy ±0.5℃, humidity accuracy ±3% RH, spore capture aperture 0.8mm, pumping rate 2L/min, can collect fungal spores in the air (such as nematode spores as vectors for pine wood nematode disease), combined with a microscope imaging module to achieve spore counting; vegetation spectral sensor (Model: ASD FieldSpec HandHeld2): spectral range 325-1075nm, resolution 3nm, sampling time 100ms, can measure leaf reflectance, and judge vegetation health status through NDVI (Normalized Difference Vegetation Index) changes (warning when the index drops ≥0.1 for 7 consecutive days and is below 0.3), discovering pest and disease stress in advance.
Deployment logic: Deploy at a density of 1 unit per 50 mu in areas with forest canopy density of 0.6-0.8. The sensor bracket height is 1.5-2m (level with the middle of the forest canopy) to avoid ground temperature and humidity interference; prioritize deployment in historical occurrence areas of pests and diseases and at the junctions of forest stands. Every 3 units form a monitoring unit, and data is cross-validated to reduce the false alarm rate (false alarm rate ≤5%); spore capture sensors need to have their capture sheets cleaned once a month, and temperature and humidity sensors need to be calibrated once a quarter to ensure data accuracy.
3. Understory Economy Scenario: Customized Sensor Combination Solutions by Crop Type
(1) Core sensors in the mushroom shed monitoring solution: Temperature and humidity sensor (Model: Sensirion SHT35): temperature accuracy ±0.2℃, humidity accuracy ±2% RH, response time 8s (humidity)/10s (temperature), protection level IP67, can withstand the high humidity environment of mushroom sheds; carbon dioxide sensor (Model: SenseAir S8-0053): measurement range 0-5000ppm, accuracy ±30ppm+3% reading, response time ≤15s, supports RS485 bus output; light intensity sensor (Model: BH1750FVI): measurement range 0-65535lux, accuracy ±20lux, response time 120ms, adjustable measurement mode (high resolution/fast mode).
Deployment and linkage: Deploy 1 temperature/humidity + light sensor combination per 20㎡, suspended 0.5m above the mushroom bags to avoid direct contact with the bags; carbon dioxide sensors are deployed on the wall opposite the ventilation outlet in the shed, at a height of 1.2m (consistent with human breathing height), with 1 unit deployed per 50㎡; all sensors are connected to the smart controller in the shed (Model: Haier HY210). When the temperature >22℃ or humidity <85%, the humidifier/ventilation fan is automatically started; when the carbon dioxide concentration >1500ppm, the skylight is linked to open, with a control accuracy of ±1℃/±5% RH.
(2) Core sensors in the understory Chinese medicinal material planting monitoring solution: Soil multi-parameter sensor (Model: Atlas Scientific EZO-EC-PH): refresh rate 1Hz, calibration cycle 6 months, chemical corrosion resistance (can withstand pH 0-14), can simultaneously measure soil EC value (accuracy ±0.1mS/cm), pH value (accuracy ±0.01pH), and temperature (accuracy ±0.5℃), probe length 8cm, salt and alkali corrosion resistance; root growth sensor (Model: CI-600 In-Situ Root Imager): uses minirhizotron technology, resolution 2 megapixels, can capture root images (once a week), and analyze changes in root length and diameter (accuracy ±0.1mm); rainfall sensor (Model: RainWise RAINEW): measurement range 0-50mm/h, accuracy ±0.2mm, tipping bucket design, supports pulse output.
Deployment logic: Soil sensors are buried at a depth of 15-20cm (the concentrated layer of Chinese medicinal material roots), with 1 unit deployed per 30㎡, avoiding fertilization areas; root growth sensors are deployed at 1 unit per mu, with the minirhizotron inserted at a 45° angle to the ground, and the pipe orifice 5cm above the ground to prevent rainwater from pouring in; rainfall sensors are deployed in open areas of the planting area, with a bracket height of 1.5m to avoid occlusion by surrounding vegetation. The data is used to adjust irrigation frequency (e.g., in Dendrobium planting areas, drip irrigation is started when the weekly rainfall <20mm).
(3) Core sensors in the understory breeding monitoring solution: Ammonia sensor (Model: Figaro TGS826): measurement range 0-100ppm, accuracy ±5ppm, response time ≤10s, built-in temperature compensation (0-50℃), can monitor ammonia concentration in the breeding area (threshold ≤15ppm); infrared body temperature monitoring camera: resolution 4 megapixels, temperature measurement range 30-45℃, accuracy ±0.3℃, can simultaneously monitor the body temperature of more than 30 poultry, automatically alarming for abnormal body temperature (e.g., chicken body temperature >42℃); water quality sensor (Model: YSI Pro2030): used for understory fish/frog breeding scenarios, can measure dissolved oxygen (accuracy ±0.1mg/L), water temperature (accuracy ±0.1℃), and conductivity (accuracy ±1%), supports 72 hours of continuous underwater monitoring.
Deployment logic: Ammonia sensors are deployed on the wall of the air outlet in the breeding area, at a height of 1.5m (poultry breathing height), with 1 unit deployed per 500㎡, coordinated with axial flow fans (automatically started when ammonia concentration >10ppm); infrared body temperature cameras are installed in the center of the top of the breeding shed, with a viewing angle covering the entire breeding area, and 1 unit deployed per 100㎡ of breeding area; water quality sensors are submerged at 1/2 the water depth of the breeding pool, with 1 unit deployed per 200㎡ of breeding pool. Data is uploaded once an hour, and the oxygen pump is started when dissolved oxygen <5mg/L.
4. Nature Reserve Scenario: Adopting a Sensor Scheme of "Animal Tracking + Environmental Perception + Human Interference Prevention and Control"
Core sensor configuration: GPS animal collars: positioning accuracy ≤2m (Beidou/GPS dual-mode), endurance 12-24 months (adjusted according to sampling frequency), waterproof level IP68, can record animal activity trajectories (sampling interval 15-60 minutes), supports remote wake-up; infrared triggered cameras (Model: Reconyx HyperFire 2 HF2X): resolution 24 megapixels, trigger time ≤0.2s, endurance ≥6 months (4 AA batteries), supports 1080P video recording (30fps), can capture wildlife images (such as giant pandas, golden monkeys), with a species recognition accuracy of ≥95%; illegal intrusion monitoring sensors: microwave detection distance 10-30m, false alarm rate ≤1%, combined with vibration cables (detection distance 50m, can identify human trampling/vehicle rolling) to form a dual intrusion early warning.
Deployment logic: GPS collars are worn by 5%-10% of the target species population (e.g., when the giant panda population is 100, 5-10 collars are worn), and the collar weight does not exceed 3% of the animal's body weight; infrared triggered cameras are deployed at wildlife corridors (such as animal paths and water sources), with the lens 0.8-1.2m from the ground, facing the direction of animal movement, deploying 1-2 units per 2km², and the viewing angles of adjacent cameras overlap by ≥10%; illegal intrusion sensors are deployed along the boundary of the reserve, with microwave sensors and vibration cables deployed alternately every 50m, combined with AI cameras (night infrared distance ≥30m) to achieve 24-hour boundary prevention and control, with an intrusion alarm response time of ≤30s.
III. Application Practice and Typical Cases of Sensor Solutions
Typical Cases in the Field of Ecological Protection
1. Natural Forest Monitoring in the Giant Panda National Park, Sichuan
Adopting a three-level monitoring system of "space-based Gaofen satellites + air-based UAVs + ground-based infrared cameras": The space-based Gaofen-6 satellite obtains vegetation coverage data of the park once every 5 days, monitoring the degradation of natural forests through NDVI index changes (threshold ±0.1); air-based UAVs equipped with RIEGL VQ-1560i LiDAR conduct full-domain scanning twice a year to build 3D models of ancient and famous trees (accuracy ±5cm), identifying key protected trees with a height >20m and DBH >80cm; ground-based deployment of 500 infrared triggered cameras, with 1 unit deployed per 2km² along animal paths and water sources, combined with GPS collars to track the activity trajectories of 20 giant pandas, achieving "vegetation-animal-environment" linked monitoring. After application, the canopy density of natural forests increased to 0.85 (an increase of 0.12 compared to 2020), the accuracy of giant panda population monitoring reached 98%, and the response time for illegal logging behavior was shortened from 48 hours to 1 hour.
2. Ecological Restoration of the Xilin Gol Grassland, Inner Mongolia
Addressing the issues of grassland desertification and forage degradation, a solution of "satellite remote sensing + soil sensors + forage height monitors" was constructed: The space-based Fengyun-3 satellite monitors changes in desertified area monthly (accuracy ±5%), triggering a warning when the proportion of desertified area >15%; ground-based Decagon 5TM soil sensors are deployed in a 1km×1km grid to real-time monitor soil moisture (threshold ≥15%vol) and salinity (EC value ≤2mS/cm) to judge the causes of grassland degradation; portable forage height meters (LY/T 635-2022 standard) conduct quarterly sampling measurements, triggering a grazing rest order when the forage height <12cm. From 2023 to 2025, the desertified area in the project area decreased by 23%, the average forage height increased from 8cm to 18cm, the carrying capacity increased from 1.2 sheep units/hectare to 2.5 sheep units/hectare, and the ecological restoration efficiency increased by 40%.
Typical Cases in the Field of Safety Prevention and Control
1. Forest Fire Prevention in the Forest Area of Lijiang, Yunnan
Deploying a monitoring network of "pyroelectric infrared + laser smoke + AI cameras" in the forest areas around Yulong Snow Mountain: FLIR Boson320 pyroelectric infrared sensors are deployed every 1000m along forest roads, combined with smoke sensors to form a 150km linear early warning defense line; Hikvision DS-2TD2617B-6/V1 AI cameras are deployed with increased density in key areas (such as tourist distribution centers and around temples), supporting intelligent recognition of open flames/smoke. The system links with the forest area fire protection base stations. When a temperature >60℃ and smoke concentration >150μg/m³ are detected, alarm information is pushed to the fire-fighting team within 30 seconds. In 2024, the system successfully warned of 17 early fire points, shortened the fire fighting response time to 15 minutes, reduced the forest fire incidence rate by 78% compared to 2022, and reduced the burned area by 85%. The construction cost of the intelligent monitoring system is about 2 million RMB, and the annual cost of traditional manual patrol is about 800,000 RMB. The cost can be recovered in the second year after the system is put into use, and the fire fighting cost is reduced by 60%.
2. Prevention and Control of Pine Wood Nematode Disease in Zhejiang
Aiming at the characteristics of fast transmission and strong concealment of pine wood nematode disease, an early warning system of "intelligent insect trapping lamp + spore capture + spectral monitoring" was constructed: In the concentrated pine forest areas of Hangzhou and Ningbo, Top Cloud-agri TPCB-II-C6.0 intelligent insect trapping lamps are deployed every 50 mu to capture Monochamus alternatus (the vector of pine wood nematode disease), and chemical control is started when the insect population density >30 heads/day; AWS620-L spore capture sensors are deployed synchronously to collect nematode spores weekly, combined with ASD FieldSpec HandHeld 2 spectral sensors to monitor the NDVI index of pine trees (warning when it drops ≥0.1). From 2023 to 2024, the infection rate of pine wood nematode disease in the project area dropped from 12% to 3.5%, the prevention and control cost was reduced by 55% compared to the traditional manual patrol mode, achieving precise prevention and control of "early discovery and early disposal."
Typical Applications in the Field of Understory Economy
1. Tremella Planting in Gutian, Fujian
Deploying Sensirion SHT35 temperature and humidity sensors (1 unit per 20㎡) and SenseAir S8 carbon dioxide sensors (1 unit per 50㎡) in tremella greenhouses, connected to the Haier HY210 smart controller to achieve automatic regulation of temperature (20±1℃), humidity (88±2% RH), and carbon dioxide concentration (1200±100ppm). When humidity <86%, the ultrasonic humidifier is linked to start; when carbon dioxide concentration >1200ppm, the ventilation fan is turned on. After application, the tremella fruiting rate increased from 85% to 98%, the yield per shed increased by 20%, the rotting rate caused by environmental fluctuations dropped from 10% to 1.5%, and the income per mu for growers increased by 12,000 RMB.
2. Dendrobium Planting in Huoshan, Anhui
Using soil multi-parameter sensors, rainfall sensors, and a drip irrigation linkage system: soil sensors are buried at a depth of 15cm to real-time monitor pH value (5.5±0.2) and EC value (0.8±0.1mS/cm). When the pH value >5.7, acidic nutrient solution is automatically drip-irrigated; rainfall sensors record weekly rainfall, and if it is <20mm, drip irrigation is started (5L/㎡ per irrigation). At the same time, CI-600 root growth sensors are deployed to track the root development of Dendrobium (generating a root growth report monthly). In 2024, the polysaccharide content of Huoshan Dendrobium increased to 38% (industry standard ≥25%), the survival rate increased from 80% to 95%, the planting cycle was shortened from 3 years to 2.5 years, and the industrial scale expanded by 30%.
3. Understory Chicken Breeding in Jiangxi
Deploying ammonia sensors (1 unit per 500㎡) and infrared body temperature cameras in the understory breeding area: when the ammonia concentration >10ppm, the axial flow fan is automatically started for ventilation; the camera real-time monitors the body temperature of the chicken flock, and when an individual with a body temperature >42℃ is found, alarm information is pushed to the breeder to isolate the sick chicken in time. Combined with water quality sensors (YSI Pro2030) to monitor the dissolved oxygen in the drinking water pool (≥5mg/L) to ensure the drinking water safety of the chicken flock. After application, the incidence of respiratory diseases in the chicken flock dropped from 15% to 3%, the survival rate increased to 98%, the marketing time was shortened by 7 days, and the breeders earned an additional 28,000 RMB per 10,000 chickens.
IV. Industrial Development Pattern and Trends of Forestry and Grassland Monitoring Sensors
Industrial Status: Global and Chinese Market Size, Division of Labor Among Major Enterprises
1. Market Size
According to Grand View Research data, the global forestry and grassland monitoring sensor market size reached 2.87 billion US dollars in 2024, and is expected to grow to 5.63 billion US dollars by 2030, with a CAGR of 11.8%. From 2024 to 2030, the global annual growth rate of forest fire prevention sensors is 10.5%, and the annual growth rate of understory economy environmental monitoring sensors is 18.2%, which are the fastest-growing segmented fields; among them, the Chinese market size accounts for about 25% (720 million US dollars). Driven by the "14th Five-Year Plan" smart forestry and grassland plan, the annual growth rate reaches 15.2%, and the growth rate of China's understory economy sensors reaches 22.5%, higher than the global average. In segmented fields, forest fire prevention sensors account for the highest proportion (32%), followed by pest and disease monitoring (25%), understory economy environmental monitoring (20%), and nature reserve animal monitoring (18%), with the remaining 5% being general meteorological/soil sensors.
2. Division of Labor Among Major Enterprises
International enterprises focus on high-end R&D and system integration, such as FLIR in the US (global market share of pyroelectric infrared sensors is 35%), RIEGL in Germany (airborne LiDAR market share is 28%), Sensirion in Switzerland (temperature and humidity sensor accuracy reaches ±0.2℃, market share in understory economy scenarios is 40%); Vectronic in Canada (GPS animal collar endurance reaches 24 months, global application rate in nature reserves exceeds 60%).
Domestic enterprises achieve breakthroughs in segmented fields. On the hardware side, represented by Top Cloud-agri (domestic market share of intelligent insect condition forecasting lamps is 55%) and Hikvision (AI forest fire prevention camera market share is 40%); on the system integration side, centered on China Forestry Group and Piesat. China Forestry Group relies on the layout of national forestry and grassland resources to provide an integrated solution of "sensors + management services," with a market share of over 80% in state-owned forest areas; Piesat, based on the PIE-Engine remote sensing cloud platform, realizes the real-time fusion of satellite data and ground sensor data, with a market share of over 60% in national park projects; start-ups such as Sengong Tech focus on special sensors for the understory economy (such as Dendrobium soil EC value sensors), occupying 30% of the regional market relying on localized calibration services.
3. Analysis of Technical Barriers
Foreign enterprises hold over 70% of patents in the fields of hyperspectral sensors and LiDAR core chips. Domestic enterprises focus on algorithm and application scenario innovation. For example, Top Cloud-agri's insect condition AI recognition algorithm has applied for 20 invention patents, and its accuracy in localized scenarios is 5-10 percentage points higher than that of similar foreign products.
Existing Problems and Challenges
1. Technical Level: Significant Gap Between Domestic and Foreign Levels
Accuracy gap: The temperature resolution of foreign pyroelectric infrared sensors reaches 0.02℃ (such as Boson640), while domestic similar products are mostly 0.05℃; the response time of foreign laser smoke sensors is ≤10s (Alphasense OPCT-10), while domestic products are mostly 20-30s, leading to a 5-10 minute delay in fire warning.
Insufficient multi-sensor collaboration capability: Foreign countries have realized the real-time fusion of space-based satellite and ground-based sensor data (such as the USFS system, with data delay ≤30s). Most domestic projects are still in the state of "data islands," and the integration cycle of space-air-ground data is as long as 2-3 days, making it difficult to support emergency decision-making.
Weak adaptability to extreme environments: Foreign sensors can withstand a wide temperature range of -40℃~85℃ (such as Campbell CS215 meteorological sensors), while domestic products are prone to data drift (error ≥2℃) below -20℃, limiting their application in low-temperature areas such as the Qinghai-Tibet Plateau and Northeast forest areas.
2. Application Level: Scenario Adaptability Needs to be Improved
Insufficient customization for the understory economy: Most sensors are general-purpose, and there are few special sensors for specific crops (such as Dendrobium and tremella). Secondary calibration is required (e.g., soil pH sensors need to adjust the measurement range to 4.5-6.5 for the Dendrobium growth environment), increasing the usage cost for farmers.
Weak grassroots operation and maintenance capability: The maintenance cycle of mountainous monitoring points is long (once a quarter), and sensor calibration relies on professionals, resulting in a data accuracy of <90% for more than 30% of grassroots stations (e.g., in some forest areas in Yunnan, the false alarm rate of smoke sensors reaches 15% due to untimely dust cleaning).
3. Policy Level: Imperfect Standards and Subsidy Systems
Lack of unified monitoring standards: The sensor parameter thresholds adopted by forestry and grassland departments in different regions are inconsistent (e.g., the smoke concentration alarm threshold for forest fire prevention is 150μg/m³ in Yunnan and 200μg/m³ in Sichuan), leading to the inability of cross-regional data to interconnect.
Limited subsidy coverage: Subsidies for understory economy sensors only cover a few pilot areas (such as Gutian, Fujian, and Huoshan, Anhui). Most farmers need to bear the sensor procurement costs themselves (3000-5000 RMB for deployment in a single shed), restricting the popularization of the technology.
Future Development Trends
1. Technical Trends: From "Single Monitoring" to "Intelligent Decision-Making"
Core sensor upgrade: Breakthroughs in wide temperature range (-40℃~85℃) and high protection (IP69K) technologies, developing special sensors for the understory economy (such as tremella humidity sensors with anti-fog coating, accuracy improved to ±1% RH); integrating AI algorithms to achieve edge computing at the sensor end (such as local recognition of animal species by infrared cameras, reducing data transmission volume by 50%).
Deep fusion of multi-source data: Constructing a "space-air-ground" real-time data link. Based on 5G/Beidou short message technology, satellite data is transmitted to ground-based monitoring stations within 30s, supporting minute-level response to emergencies such as fires and pests.
Low-cost breakthrough: Developing sensors based on MEMS technology. The cost of MEMS sensors is reduced by 70% compared to traditional sensors. For example, the price of a single soil moisture sensor drops to 80 RMB, and the deployment cost for a single shed for farmers drops from 5000 RMB to 1500 RMB. Combined with a 30% agricultural subsidy, the actual investment is only 1050 RMB, promoting the popularization rate from 15% to 50%.
2. Application Trends: From "Ecological Protection" to "Industry Empowerment"
Whole industry chain monitoring of forestry and grassland: Extending sensor applications to the processing and circulation links of forest products (such as wood moisture sensors monitoring storage moisture to avoid mildew; understory medicinal material traceability sensors recording growth environment data to enhance product premium).
Lightweight applications for farmers: Launching a "sensor + mini-program" model, where farmers receive environmental warnings via WeChat (such as soil pH anomaly reminders in Dendrobium planting areas), supported by on-site calibration services to lower the operation and maintenance threshold.
3. Policy Trends: Strengthening Standards and Ecological Support
Unified monitoring standards: Led by the National Forestry and Grassland Administration, formulate the Technical Specification for Forestry and Grassland Monitoring Sensors, clarify the parameter thresholds of sensors in various scenarios (such as unifying the smoke concentration for forest fire prevention to 150μg/m³) and data interface standards to achieve national data interconnection.
Expanding the scope of subsidies: Include understory economy sensors in the agricultural subsidy catalog, providing 30%-50% subsidies for farmers' procurement; establish a special fund for forestry and grassland sensor R&D to support domestic enterprises in breaking through core technologies (such as wide temperature range sensors and AI edge computing modules).
Building an industrial ecosystem: Promoting cooperation among "sensor enterprises + forestry and grassland cooperatives + research institutions," such as Top Cloud-agri jointly developing special sensors with the Fujian tremella cooperative, and the Chinese Academy of Forestry providing calibration technical support, forming a closed loop of "R&D - application - feedback."
Conclusion
Forestry and grassland industry monitoring is a key grasp for ensuring ecological security and promoting rural revitalization. As the "perceptual nerves," intelligent sensors have achieved full-scenario coverage from ecological protection to the understory economy. By constructing a "space-air-ground" integrated monitoring system, it can accurately capture vegetation changes, fire hazards, pest and disease dynamics, and crop growth environment data, providing a scientific basis for forestry and grassland resource management.
Currently, although the domestic forestry and grassland monitoring sensor industry has achieved breakthroughs in localized applications, there is still a gap with the international leading level in core accuracy, extreme environment adaptability, and multi-source data fusion capability.
In the future, focus needs to be placed on three major directions: First, technical breakthroughs, tackling wide temperature range, high-precision sensors, and AI edge computing modules; second, scenario deepening, developing special sensors for the understory economy, and promoting lightweight applications; third, ecosystem construction, improving unified standards and subsidy policies, and forming a closed loop of "R&D - application - feedback". Through the deep integration of sensor technology and the forestry and grassland industry, a win-win situation for both ecological and economic benefits can be realized, providing solid support for building a Beautiful China and promoting green development.
Author | Guo Yuansheng, Deputy Director of the Science and Technology Committee of the Central Committee of the Jiusan Society, Executive Vice Chairman of the China Sensor and IoT Industry Alliance
Editor | Yang Pengyue Art Editor | Ma Liya Supervisor | Zhao Chen