AU2017376837A1

AU2017376837A1 - Detection of environmental conditions

Info

Publication number: AU2017376837A1
Application number: AU2017376837A
Authority: AU
Inventors: Anthony Clark; Jason CREAN; Ian Mcgowen
Original assignee: Crown In Right Of State Of New South Wales Acting Through Department Of Primary Indu
Current assignee: Crown In Right Of State Of New South Wales Acting Through Department Of Primary Industries An Office Of Nsw Department Of Industry Skills And Regional Development
Priority date: 2016-12-16
Filing date: 2017-12-15
Publication date: 2019-07-25
Also published as: AU2016273991A1; WO2018107245A1

Abstract

An environmental monitoring system (1) and method (100) for detecting an environmental condition of a specified area (3). The system (1) include one or more remote sensors (4) and/or climate data source (9) to provide sensor data indicative of soil water content, plant growth and rainfall associated with the specified area. A processing device (13) is configured to: determine (110) soil water content, plant growth and rainfall based on sensor data and historical data; determine (120) a drought direction associated with the specified area (3) that indicates a trend in precipitation; and determine (130) indicators for soil water content, plant growth and rainfall during drought conditions and normal conditions. The processing device further determines an environmental condition based on: (i) the drought direction; and (ii) the determined soil water content, plant growth and rainfall with the determined respective indicators. Based on the determined environmental condition, a notification is sent (140) indicating the environmental condition.

Description

Detection of environmental conditions

Technical Field [0001] The present disclosure relates to an environmental monitoring system and method of detecting environmental conditions of a specified area.

Background [0002] Environmental conditions can have an impact on the health of living organisms in a particular geographic area. For farmers, periods of drought can affect the productivity of their land. Historically the term “drought” may have multiple definitions and be defined, or affected, by one or more of meteorological, hydrological, agronomic and social-economic phenomenon.

[0003] Monitoring environmental conditions such as drought can assist stakeholders such as farmers and governments prepare for, and manage resources for these conditions. Known monitoring systems and methods may provide general environmental conditions on a larger regional scale, but lack the granularity to provide stakeholders at their respective areas information that is accurate to their local area of interest. As an example, a region may be in severe drought, but the impact may be lower along the banks of a river passing through that region.

[0004] Any discussion of documents, acts, materials, devices, articles or the like which has been included in the present specification is not to be taken as an admission that any or all of these matters form part of the prior art base or were common general knowledge in the field relevant to the present disclosure as it existed before the priority date of each claim of this application.

[0005] Throughout this specification the word comprise, or variations such as comprises or comprising, will be understood to imply the inclusion of a stated element, integer or step, or group of elements, integers or steps, but not the exclusion of any other element, integer or step, or group of elements, integers or steps.

WO 2018/107245

PCT/AU2017/051403

Summary [0006] An environmental monitoring system for detecting an environmental condition of a specified area, the system comprising: one or more remote sensors and/or a climate data source to provide sensor data indicative of soil water content, plant growth and rainfall associated with the specified area; a database to store historical data based on historical soil water data, historical plant growth data, and historical rainfall data; and a processing device. The processing device is configured to: determine soil water content, plant growth and rainfall based on sensor data from the one or more remote sensors and/or climate data source and historical data; determine a drought direction associated with the specified area that indicates a trend in precipitation based on the determined rainfall and historical rainfall data; determine respective indicators for soil water content, plant growth and rainfall during drought conditions and normal conditions associated with the specified area that are based on historical data; and determine a warning environmental condition associated with the specified area. Determining a warning environmental condition is based on determination of:

- a drought direction indicating a trend towards precipitation deficit; and at least one of:

- determined soil water content is greater than the indicator for soil water content during drought conditions but less than the indicator for soil water content during normal conditions;

- determined plant growth is greater than the indicator for plant growth during drought conditions but less than the indicator for plant growth during normal conditions; and

- determined rainfall is greater than the indicator for rainfall during drought conditions but less than the indicator for rainfall during normal conditions.

[0007] Based on determining a warning environmental condition, the processing device is further configured to send, over a communications network, a notification to a communication device indicating a warning environmental condition associated with the specified area.

WO 2018/107245

PCT/AU2017/051403 [0008] In the system, the processing device may be further configured to: determine the indicators for soil water content, plant growth and rainfall during early recovery conditions associated with the specified area based on historical data; and determine an early recovery environmental condition associated with the specified area. Determining an early recovery environmental condition may be based on determination of:

- a drought direction indicating a trend towards precipitation surplus; and

- determined soil water content is within the indicator for soil water content during early recovery conditions; and

- determined plant growth is within the indicator for plant growth early recovery conditions; and

- determined rainfall is within the indicator for rainfall during early recovery conditions.

[0009] Based on determining an early recovery environmental condition, the processing device may be further configured to send, over a communications network, a notification indicating an early recovery environmental condition associated with the specified area.

[0010] In the system, the processing device may be further configured to: determine the indicators for soil water content, plant growth and rainfall during strong recovery conditions associated with the specified area based on historical data; and determine a strong recovery environmental condition associated with the specified area. Determining a strong recover environmental condition may be based on determination of:

- a drought direction indicating a trend towards precipitation surplus; and

- determined soil water content is within the indicator for soil water content strong recovery conditions; and

- determined plant growth is within the indicator for plant growth strong recovery conditions; and

WO 2018/107245

PCT/AU2017/051403

- determined rainfall is within the indicator for rainfall during strong recovery conditions.

[0011] Based on determining a strong recovery environmental condition, the processing device may be further configured to send, over a communications network, a notification indicating a strong recovery environmental condition associated with the specified area.

[0012] In the system, the processing device may be further configured to: determine a drought environmental condition associated with the specified area based on determination of at least one of:

- determined soil water content is less than the indicator for soil water content during drought conditions;

- determined plant growth is less than the indicator for plant growth during drought conditions; and

- determined rainfall is less than the indicator for rainfall drought conditions.

[0013] Based on determining a drought environmental condition, the processing device may be further configured to send, over a communications network, a notification indicating a drought environmental condition associated with the specified area.

[0014] In the system, the processing device may be further configured to: determine a non drought environmental condition associated with the specified area based on determination of:

- determined soil water content is within the indicator for soil water content during normal conditions; and

- determined plant growth is within the indicator for plant growth during normal conditions; and

- determined rainfall is within the indicator for rainfall during normal conditions.

WO 2018/107245

PCT/AU2017/051403 [0015] Based on determining a non drought environmental condition, the processing device may be further configured to send, over a communications network, a notification indicating a non drought environmental condition associated with the specified area.

[0016] Thus the system may provide notifications to stakeholders (such as farmers) more accurate information indicative of the current state of their specified area of interest as well as the trend. For example, where the environmental conditions are between normal conditions and drought conditions it may be difficult for the stakeholder to objectively determine the trend (i.e. are conditions deteriorating further into drought or recovering). The present system provides an indicator that captures the trend. This may allow the stakeholders to more effectively act and prepare for present and future conditions.

[0017] Furthermore, previous systems of drought monitoring may include looking at climate at the larger level and area (e.g. meteorological monitoring run by national or state level organisations) which, although may be temporally relevant to farmers (due to resources to provide regular updates to weather forecast), lack resolution, granularity and accuracy. In particular, such monitoring may not take into account local conditions that may be relevant. On the other hand, farmers may have various sensors and apparatus to monitor some factors indicative of the local conditions, this may be on an ad-hoc basis and such tests may be deficient in factoring other variables (because of lack of resources, coordination of information with other farmers or organisations, etc.). The present disclosure may ameliorate or overcome these issues.

[0018] In some example, the processing device may be further configured to determine the soil water content, plant growth and rainfall as an aggregation of respective sensor data from a specified preceding period.

[0019] In the system, the processing device may be further configured to:

- determine a range or threshold value as the indicator for soil water content in normal conditions;

- determine a range or threshold value as the indicator for plant growth in normal conditions;

WO 2018/107245

PCT/AU2017/051403

- determine a range or threshold value as the indicator for rainfall in normal conditions;

- determine a range or threshold value as the indicator for soil water content in drought conditions;

- determine a range or threshold value as the indicator for plant growth in drought conditions; and

- determine a range or threshold value as the indicator for rainfall in drought conditions.

[0020] In the system, the processing device may be further configured to:

- determine a range or threshold value as the indicator for soil water content in early recovery conditions;

- determine a range or threshold value as the indicator for plant growth in early recovery conditions;

- determine a range or threshold value as the indicator for rainfall in early recovery conditions;

- determine a range or threshold value as the indicator for soil water content in strong recovery conditions;

- determine a range or threshold value as the indicator for plant growth in strong recovery conditions; and

- determine a range or threshold value as the indicator for rainfall in strong recovery conditions.

[0021] The range or threshold value may be expressed as a percentile range or percentile threshold value.

WO 2018/107245

PCT/AU2017/051403 [0022] The processing device may be further configured to determine the soil water content, plant growth and rainfall as normalised data values based on historical soil water content, historical plant growth and historical rainfall.

[0023] The system may be further configured to detect an environmental condition of a geographic parish that includes an associated plurality of specified areas, wherein the processing device is further configured to:

- determine an environmental condition for each of the associated plurality of specified areas;

- determine the environmental condition associated with the parish based on the environmental condition that occurs most frequently in the associated plurality of specified areas, wherein to send, over the communications network, the notification to the communication device includes the processing device configured to send the environmental condition associated with the geographic parish.

[0024] In the system, the associated plurality of specified areas may include specified areas that the geographic parish overlap.

[0025] In the system, the associated plurality of specified areas may further include specified areas that are proximal to the boundary of the geographic parish.

[0026] In the system, the one or more remote sensors may comprise one or more of the following:

- a soil water sensor to provide soil water data associated with the specified area;

- a rain gauge to provide rainfall data associated with the specified area; and

- a plant growth monitoring device to provide plant growth data associated with the specified area.

WO 2018/107245

PCT/AU2017/051403 [0027] In the system, the one or more remote sensors may comprise one or more of the following:

- a temperature sensor;

- a light sensor;

- an evaporation gauge;

- a humidity sensor;

- an anemometer; and

- a barometer.

[0028] The system may further include one or more aerial drones, wherein the aerial drones have one or more of the remote sensors on board. The aerial drones may include a communication module to send data to the processing device.

[0029] In the system, the climate data source may be associated with one or more of the remote sensors that collect sensor data associated with the specified area.

[0030] In the system, the processing device is further configured to generate, at a display, a graphical representation of the notifications as an overlay of a map that includes at least the specified area.

[0031] A computer-implemented method for detecting an environmental condition of a specified area, the method comprising:

- determining soil water content, plant growth and rainfall based on sensor data from one or more remote sensors and/or climate data source and historical data, wherein the one or more remote sensors and/or climate data source provides sensor data indicative of soil water content, plant growth and rainfall, and

WO 2018/107245

PCT/AU2017/051403 wherein the historical data is based on at least historical soil water data, historical plant growth and historical rainfall data;

- determining a drought direction associated with the specified area that indicates a trend in precipitation based on the determined rainfall and historical rainfall data;

- determining respective indicators for soil water content, plant growth and rainfall during drought conditions and normal conditions associated with the specified area that are based on historical data;

- determining a warning environmental condition associated with the specified area based on determination of:

- determined rainfall is greater than the indicator for rainfall during drought conditions but less than the indicator for rainfall during normal conditions;

- based on determining a warning environmental condition, sending, over a communications network, a notification to a communication device indicating a warning environmental condition associated with the specified area.

[0032] The method may further comprise:

WO 2018/107245

PCT/AU2017/051403

- determining the indicators for soil water content, plant growth and rainfall during early recovery conditions associated with the specified area based on historical data;

- determining an early recovery environmental condition associated with the specified area based on determination of:

- a drought direction indicating a trend towards precipitation surplus; and

- determined rainfall is within the indicator for rainfall during early recovery conditions; and

- based on determining an early recovery environmental condition, sending, over a communications network, a notification indicating an early recovery environmental condition associated with the specified area.

[0033] The method may further comprise:

- determining the indicators for soil water content, plant growth and rainfall during strong recovery conditions associated with the specified area based on historical data;

- determining a strong recovery environmental condition associated with the specified area based on determination of:

- a drought direction indicating a trend towards precipitation surplus; and

WO 2018/107245

PCT/AU2017/051403

- determined rainfall is within the indicator for rainfall during strong recovery conditions; and

- based on determining a strong recovery environmental condition, sending, over a communications network, a notification indicating a strong recovery environmental condition associated with the specified area.

[0034] The method may further comprise:

- determining a drought environmental condition associated with the specified area based on determination of at least one of:

- determined rainfall is less than the indicator for rainfall drought conditions; and

- based on determining a drought environmental condition, sending, over a communications network, a notification indicating a drought environmental condition associated with the specified area.

[0035] The method may further comprise:

- determining a non drought environmental condition associated with the specified area based on determination of:

WO 2018/107245

PCT/AU2017/051403

- determined rainfall is within the indicator for rainfall during normal conditions; and

- based on determining a non drought environmental condition, sending, over a communications network, a notification indicating a non drought environmental condition associated with the specified area.

[0036] The method may further comprise generating, at a display associated with the communications device, a graphical representation of the notification as an overlay on a map that includes at least the specified area.

[0037] Software that, when installed on a computer, causes the computer to perform the method described above.

Brief Description of Drawings [0038] Examples of the disclosure will now be described with reference to the figures below:

[0039] Fig. 1 is a schematic of a system for detecting environmental conditions;

[0040] Fig. 2 is a flow diagram of a method performed by a processing device in the system;

[0041] Fig. 3 is a diagram of an example phases of wet and dry conditions;

[0042] Figs. 4a to 4d are additional examples of wet and dry conditions;

[0043] Fig. 5 is a schematic example of determining soil water content;

[0044] Fig. 6 is a schematic of an optimisation workflow for determining model and model parameters;

WO 2018/107245

PCT/AU2017/051403 [0045] Fig. 7 is a graph illustrating verification of the model and model parameters;

[0046] Fig. 8 is a map of a region including notification of environmental conditions overlayed;

[0047] Fig. 9 is shows a map of a subset of a region that includes a geographic parish, where the geographic parish overlaps a plurality of gridded specified areas;

[0048] Fig. 10 illustrates a user interface at a display showing a map of a state divided into regions with notification of environmental conditions overlayed;

[0049] Fig. 11 illustrates the user interface at the display after selection of a region and showing shires within the region;

[0050] Fig. 12 illustrates the user interface at the display after selection of a shire and illustrating details of a selected geographic parish in the selected shire;

[0051] Fig. 13 illustrates a schematic of one variation of the system; and [0052] Fig. 14 illustrates a schematic example of a processing device.

Description of Embodiments

Overview [0053] The present disclosure includes an environmental monitoring system 1 for detecting environmental conditions of a specified area 3 as illustrated in Fig. 1. The system 1 includes remote sensors 4 that detect information associated with the specified area 3, and sends respective data to a processing device 13. A climate data source 9 also sends data associated with the specified area 3 to the processing device 13.

[0054] The remote sensors 4 may include rain gauges 7, soil water sensors 5, temperature sensors, plant growth monitoring devices, and other sensors. The climate data source 9 may include meteorological data sources and sensors that provide rainfall data. In some examples,

WO 2018/107245

PCT/AU2017/051403 the climate data source 9 may be, or receive, data from remote sensors 4, such as rain gauges, associated with the specified area 3.

[0055] A database 11 stores historical data that is based on historical information on soil water, plant growth and rainfall associated with the specified area.

[0056] As illustrated in Fig. 2, the processing device 13 receives data from the remote sensors 4 and climate data source 9 to determine 110 the soil water content, plant growth and rainfall associated with the specified area 3. The processing device 13 further determines 120 a drought direction associated with the specified area 3 that indicates a trend in precipitation based on the rainfall data from the climate data source 9 and the historical rainfall from the database 11. The processing device 13 may also determine 130 respective indicators for soil water content, plant growth, and rainfall during conditions such as drought conditions, normal conditions, early recovery conditions, strong recovery conditions. Such indicators may include threshold values. This may be based on historical data associated with the specified area 3.

[0057] The processing device 13 may then determine 140 one or more environmental conditions based on the drought direction, soil water, plant growth, rainfall and the determined indicators. Such environmental conditions may include one of:

- a warning 33 environmental condition;

- an early recovery 37 environmental condition;

- a strong recovery 39 environmental condition;

- a drought 35 environmental condition; and

- a non-drought 31 environmental condition.

[0058] An example of these environmental conditions during phases of dry and wet conditions is illustrated in Fig. 3.

WO 2018/107245

PCT/AU2017/051403 [0059] The processing device 13 may then send, over a communications network 15, a notification to a communication device 17 indicating the determined environmental condition associated with the specified area 3. An example of a communication device 17, may be a mobile communication device or a computer of a farmer who is then alerted to the environmental condition their property and may, in turn, take appropriate action to prepare for, take advantage or and/or mitigate the effects of the environmental condition. Since this notification is for the specified area 3, this may provide the farmer (or other stakeholder) more granular and accurate information than other systems.

[0060] One technical advantage of the system 1 is that it can provide an indication on whether conditions are deteriorating or improving at parts of the phase between drought and non-drought. This is illustrated in Fig. 3 whereby the “warning” and “strong recovery” on the curve may have similar absolute wet/dry characteristics but should be categorised as different environmental conditions to take into account a trend in improving or deteriorating conditions. This is in contrast to the technical problem of other systems that may show levels of absolute wet or dry characteristics without indicating deterioration or improvement. Such information may be of significant importance to stakeholders such as farmers and governments as they can use this as basis to allocate resources, such as water, feed for livestock, equipment, etc. This may also determine other actions including control of irrigation, planting, fertilising, harvesting, or implementation of drought protection measures. In some examples, this may include using the notification as an input to an automated system such as automated watering and feeding systems based on the environmental condition.

[0061] Furthermore, it is to be appreciated that the phases of dry and wet conditions, in the real world, do not occur in consistent and regular cycles that may be suggested in Fig. 2. Figs. 4a to 4d show other examples of phases where the warning and recovery take various shapes. The system may advantageously provide an indication of the environmental condition of the specified area 3 in such other, non-idealised, phases of variability.

[0062] The parts of the system 1 will now be described in detail.

WO 2018/107245

PCT/AU2017/051403

The remote sensors 4 [0063] The remote sensors 4 may be used to collect sensor data used to determine soil water data, plant growth data and rainfall data associated with the specified area 3. It is to be appreciated that in some examples, the specified area 3 may have the remote sensors 4 located therein to provide direct measurements and data from the specified area.

[0064] However it is to be appreciated that in some examples, a specified area 3 may not have sensors 3 located therein and therefore sensor data associated with the specified area may be determined by interpolating data directly measured from other areas. For example, if a type of remote sensor 4 is not physically located in the specified area 3, but adjacent areas have such remote sensors 4, then the information from such remote sensors may provide data that can be used for determining the environmental condition of the specified area 3. In one example, say the specified area 3 is in between two other areas that have remote sensors, the interpolated value) may be used to provide that sensor data for the specified area 3.

[0065] The remote sensors 4 may include individual sensors that are connected to the processing device 13 via a communications network 15. In other examples, the sensors may send data to a data logger, which in turn, is connected to the communications network 15. In some other examples, the remote sensors 4 may be part of a meteorological (weather) station.

[0066] In some further examples, the remote sensors 4 may include sensors operated by, or with the permission of, property owners or lessors. For example, this may include remote sensors 4 used by farmers on the property in the specified area 3. In some examples, from sent from such sensors are aggregated to the specified area 3 (or larger area) to reduce privacy or confidentiality concerns. Therefore the remote sensors 4 may include a network of sensors from various including individuals, business, organisations, or other stakeholders and service providers that in other circumstances may not wish to share data or collaborate with one another.

[0067] Examples of remote sensors 4 that may be used to determine soil water content, plant growth and rainfall will now be described. It is to be appreciated that this determination may include using data from such remote sensors together with models in the processing device 13.

WO 2018/107245

PCT/AU2017/051403

Rain gauge [0068] The remote sensors 4 may include rain gauges 7 to determine rainfall associated with the specified area 3. Various types of rain gauges may be used including a tipping bucket rain gauge that tips after a specified amount of precipitation is collected, and wherein the numbers of tips are counted to determine the total precipitation. Other examples may include a rain gauge based on weighing the collected precipitation.

[0069] The rain gauge may provide rainfall data of the specified area as well as providing data that may be used by models to determine the soil water content and plant growth.

Soil water sensor [0070] Soil water sensors 5 provide data that can assist in determining water content in the soil. A known method of determining soil water content may include gravimetric analysis of a soil sample, which as an example may include obtaining and weighing the soil sample, drying the soil sample and weighing the dried soil sample. The weight difference may then be used to determine the water content of the soil sample.

[0071] However such a method may not be practical in some circumstances as this may be labour intensive or slow and therefore undesirable for medium to long term monitoring. Other methods for determining (which includes estimating) the water content may include using a soil capacitance probe. Changes to the water content of the soil can vary the capacitance and this change can be measured to determine changes (or the value) of the water content.

[0072] An example of a soil water sensor includes the OTTO family of soil moisture probes offered by TOIP Pty Ltd. Another example includes the AQUACHECK SUB-SURFACE PROBE offered by AQUACHECK (PTY) LTD.

[0073] It is to be appreciated that the soil type, composition, crop, etc. may affect the readings from the soil water sensors. Therefore the sensors may be calibrated by using other methods of determining soil water content. For example, the soil capacitance probe may be calibrated by measuring soil capacitance with a soil sample, whereby the soil sample further undergoes gravimetric analysis to determine the water content.

WO 2018/107245

PCT/AU2017/051403 [0074] It is to be appreciated that in addition to sensors that take measurements from the soil of the specified area 3, data from other sensors may be used to infer the soil water content. These may include information from rain gauges, temperature sensors, evaporation sensors, radiometers (e.g. light sensors), etc. Such data may be used in models (discussed in further detail below) to determine the soil water content in the soil.

Temperature sensor [0075] The remote sensor 4 may also include a temperature sensor, such as a thermometer, associated with the specified area 3. This may include a digital thermometer located at the specified area 3 to measure air and/or soil temperature that is sent, over the communications network 15, to the processing device 13.

[0076] In other examples, remote sensors may include electromagnetic sensors in the visible and non-visible ranges (such as the infrared spectrum). In one example, thermography system may be used to determine temperatures associated with the specified area 3, and may include a heat map of a geographical region. For example, a thermography camera may be mounted on observation aircraft, aerial drones 16 or satellites 18.

[0077] Temperature data may be used in models to determine soil water content and plant growth in the specified area. For example, the temperature (including maximum temperature, minimum temperature, average temperature, median temperature) may be used to assist in determining evaporation and hence the soil water content in the specified area. This may be further based on historical data to determine the correlation and parameters for the model.

[0078] Similarly the temperature may also be used to model the plant growth, with the temperature as factors that affects plant growth, such as regulation of photosynthesis, crop development, frost and decay of plant material.

Light sensor [0079] The remote sensors may also include a light sensor to determine solar radiation associated with the specified area 3. In some examples, this may include a radiometer to measure the radiant flux of sunlight. In other examples, this may include a pyranometer. In some examples, this may be located at the specified area 3. In other examples, this may

WO 2018/107245

PCT/AU2017/051403 include sensors away from the surface of the specified area 3, such as from an aircraft, aerial drone 16 or satellite 18.

Evaporation gauges [0080] The remote sensors 4 may also include an evaporation gauge to assist determination of evaporation in the specified area 3. The evaporation gauge may include an evaporation pan with water, whereby the change in level of water in the pan may be used to determine evaporation in the specified area 3. The evaporation rate may be used in a model to determine the amount of water evaporation in the soil, and hence assist determination of the soil water content. As the amount of water also affects plant growth, this may also be used to model plant growth.

Humidity sensor/ Hygrometer [0081] The remote sensors 4 may also include hydrometer(s) to measure moisture content in the atmosphere associated with the specified area 3. The hydrometer may use one or more changes in capacitance, resistance and/or thermal conductivity of air to determine the humidity.

Anemometer [0082] The remote sensors 4 may also include anemometers to measure wind speed associated with the specified area 3. The wind speed may be used to provide parameters for the models that relate to convection that may affects evaporation and plant growth.

[0083] The wind speed (and direction) may also be used to determine changes in temperature at the specified area 3, or other areas. For example, the direction and speed of air (having a temperature) may be modelled to determine the temperature of a specified area 3 downwind.

WO 2018/107245

PCT/AU2017/051403

Barometers [0084] The remote sensors 4 may also include barometers to measure pressure associated with the specified area 3. This may be used to assist modelling climate conditions that affect the specified area.

Plant growth monitoring device and sensors [0085] Examples of data that is collected from remote sensors 4 that may be used as indicators of plant growth include:

• Advanced Very High Resolution Radiometer (AVHRR) Normalised Difference Vegetation Index (NDVI) - this determines the photosynthetic capacity of vegetation based on the reflected radiation in the visible and near-infrared wavelengths.

• Moderate Resolution Imaging Spectroradiometer (MODIS) Leaf Area Index (LAI) This measures the number of equivalent layers of leaves relative to a unit of ground area;

• MODIS Fraction of Photosynthetically Active Radiation (FPAR) - This is a measure of the proportion of available radiation in the photosynthetically active wavelengths that are absorbed by a canopy.

[0086] These plant growth monitoring devices and systems provide plant growth data that may be used to determine plant growth This may include evaluating the photochemical reflectance index to assist in determining plant growth. This may include a system that uses such plant growth data in a model (discussed in further detail below) to determine the plant growth. This may include evaluating the photochemical reflectance index to assist in determining plant growth.

[0087] It is to be appreciated that other sensors or methods may be used to receive indicators of plant growth. For example, paddock monitoring with such as active optic techniques (for example the GreenSeeker (hand held or vehicle mounted) product offered by TRIMBLE INC. that emit light towards target plants and measures reflected light), visual (RGB) cameras or Mutispectral and Hyperspectral sensors for determining the crop and pasture yield that may be indicative of plant condition in the specified area 3.

WO 2018/107245

PCT/AU2017/051403 [0088] Further information about crop and pasture growth may be determined by a variety of non-destructive and or destructive methods. Examples of non-destructive methods include standardised visual estimation of pasture and crop biomass, physical measurements of plant height with a ruler, laser and sonar device and quadrant based counts of pasture species composition (BOTANAL). Destructive methods include harvesting of crop and pasture biomass for weight determination and or laboratory analysis.

Remote sensor platforms [0089] In some examples, one or more of the remote sensor(s) may be mounted on mobile platforms. This may allow the remote sensors 4 to be easily shared with multiple specified areas compare to permanent or semi-permanent sensors.

[0090] This may include aircraft, ground vehicles, satellites 18 etc. This may also include mobile remote sensor stations that are containerised and/or trailer mounted. In further examples, one or more of the remote sensors may be mounted on an aerial drone. Such aerial drones may be autonomous, semi-autonomous or remotely piloted. Aerial drones 16 may allow the remote sensors 4 to be more mobile and cover a greater area compared to fixed sensors whilst being more cost effective that manned aircraft and satellites. Furthermore, aerial drones 16 may safely fly lower than manned aircraft and satellites 18 which may allow the drones to captures data at a higher resolution. Therefore, the higher resolution data may be processed to allow individual farmers to monitor environmental conditions to a lower level - such as a paddock level or lower. In some examples, the data from the aerial drones may be private data for the farmer that is used in conjunction with the environmental monitoring system 1 to determine environmental conditions for specific areas of their land. In some examples the aerial drones may include a communication module to provide real-time or near real-time data.

Climate data source 9 [0091] The remote sensors 4 may include sensors utilised by stakeholders and others in the specified area 3. However it is to be appreciated that some remote sensors 4 may be operated by government, university, or other research organisations such as meteorological organisations. Such organisations may receive information from a large number of remote

WO 2018/107245

PCT/AU2017/051403 sensors 4 and combine them to provide climate data for an area. In some examples, such remote sensors may include weather radar and meteorological sensors on board aircraft, aerial drones 16 and/or satellites 18. The information may therefore be compiled and stored at a meteorological bureau as a climate data source (9).

[0092] Thus in some examples of the system 1, the processing device 13 may receive at least some of the data indirectly from the remote sensors 4, as the data is procured from the climate data source (9).

Database 11 [0093] The database 11 may receive data from various sources, including remote sensors 4, farmers, stakeholders, weather bureau, other organisations, and the processing device. The database 11 may store historical data based on sensor data received from remote sensors 4 in the past. This may include historical data in relation to soil water data, plant growth data, and rainfall data. In some examples, the historical data may also be based on historical records from other sources, such as historical records of a meteorological bureau, data from other research (such as from universities, scientists or other organisations). Forecast data may also be received by the data base from meteorological organisations like the Bureau of Meteorology, CSIRO, the Hadley Centre, the National Centre for Atmospheric Research and others.

[0094] In some examples the database 11 may be distributed and include multiple databases 11 that are in communication with the processing device 13. For example, categories of data may be stored separately based on the respective information source (e.g. historical rainfall data may be stored at the meteorological bureau). This uses Application Programming Interfaces, high speed data transfer platforms (e.g. Aspera transfer platform offered by ASPERA INC.) and standard File Transfer Protocols. It is to be appreciated that the historical data may also be stored on cloud storage.

WO 2018/107245

PCT/AU2017/051403

Processing device 13 [0095] The processing device 13 may be operated centrally, or distributed to multiple processing devices across the network 15. An example of a processing device is discussed in further detail below.

[0096] The processing device 13 performs the method 100 shown in Fig. 2 of: determining

110 the soil water content, plant growth and rainfall; determining 120 the drought direction; and determining 130 indicators for soil water content, plant growth and rainfall during various conditions such as drought, non drought and recovery conditions. Furthermore, the processing device 13 evaluates these factors to determine 140 the environmental condition. In response to determining the environmental condition, the processing device 13 sends 150 notifications to communication devices that may be used to alert stakeholders affected by the environmental condition.

[0097] In some examples, the notifications may include generating, at a communications device 17, a map overlay showing specified area(s) affected by one, or more, environmental condition.

[0098] In some examples, the database 11 may be collocated with the processing device 13, whereby the system 1 is operated by an organisation such as a government agency or primary industries organisation.

[0099] The method performed by the processing device 13 will now be described in detail.

Determining 110 the soil water content, plant growth, rainfall [0100] Determining 110 the soil water content, plant growth, rainfall and drought direction will now be described individually.

Soil water content [0101] The soil water content is indicative of water in the soil which is an indicator of the water readily available to plants in the specified area 3. For convenience in calculation and expression, in some examples the soil water content may be normalised and expressed as a

WO 2018/107245

PCT/AU2017/051403 percentile compared to historical soil water content. For example, expressing the soil water content between 1-10 may indicate that there is little or no water in the soil, whereas expressing the soil water content close to 100 may indicate it is close to the historical maximum amount of moisture in the soil.

[0102] In some examples, the soil water content may also be expressed as a summation of the soil water content over a preceding time period. For example, as a summation of a preceding twelve month period. This summation may provide a better general trend of the soil water content of the specified area 3 over the preceding time period as opposed to a snapshot at a specific point in time. The plant growth and rainfall discussed below may also be expressed as a summary and normalised in a similar manner.

[0103] The soil water content will differ between layers of soil. For example, the very top layer of soil may be drier after sunshine than lower layers. In some examples, the relevant layers of soils are layer 1 (0 to 10 centimetres from the top) and layer 2 (11 to 45 centimetres from the top). Thus the determination of water content may include an aggregation of these two layers. However, in it is to be appreciated that some stakeholders, due to their requirements, find other layers of soil relevant. For example, a farmer may have crop that has shallow roots such that the soil water content of layer 1 is more important.

[0104] In some examples, the soil water may be measured directly by soil water sensors measuring the soil. However in some circumstances this may not be practical to measure on a daily basis due to time, required number of samples, and volume of data.

[0105] Therefore in some examples, the soil water content is determined based on receiving sensor data that is then input into a model to determine the soil water content for the specified area 3. From time to time, this model may be calibrated using data directly measured from the soil water sensor(s) 5 to improve accuracy.

[0106] An example of determining the soil water content with a model will now be described.

[0107] The inputs to the model may include rainfall, temperature, humidity, winds and other meteorological variables to derive potential evapotranspiration.

WO 2018/107245

PCT/AU2017/051403 [0108] Referring to Fig. 5, the model 61 may include determining the soil water content at respective layers based on:

- Soil parameters (which may be based on historic values);

- Rainfall (at the specified area);

- Infiltration of water into the soil of the specified area 3 and drainage of water out of the specified area);

- Runoff of water from the specified area

- Green and dead plant cover (that may reduce evaporation and/or capture moisture from condensation);

- Transpiration (the water drawn from the soil by the plant)

- Soil evaporation (the water evaporated at the soil surface) [0109] Essentially, the model takes into account factors that affect the soil water content to provide a value of the soil water content for the specified area.

[0110] In one example, the runoff may be determined as:

Runoff = Cover_term x (Rain -(1--_{χ SWdef}._c.^

Rain_intensity = max(100, Rain x int_intercept + int_slope x time_op _year) , 2 x pi x (day + 15) time_Of_year cos( )

365

CoveTf_:erm — (1 Run_cover) tsdm^runoff-^power

RllTlcover ~

(tsdm^runo'^FV°^wer + yld_cover50 ^runoffp°^wer)

WO 2018/107245

PCT/AU2017/051403 where

Run_cover\s the fractional cover modifier on the process of runoff tsdm is total standing dry matter yldcover50 is the biomass value at which the rate of runoff begins to decline runoff_powetw> a parameter governing the shape of the runoff-cover relationship [0111] In one example, the infiltration and drainage may be determined as:

Infiltratel = Rain - Runoff

Drainage 1 = max(0, SW1 + Infiltratel -FC1)

SWl-min(FCl, SW1 + Infiltratel)

Drainage2 - max(0, SW2 + Drainagel - FC2)

SW2=min(FC2, SW2 + Drainagel)

Drainage3 = max(0, SW3 + Drainage2 - FC3)

SW3=min(FC3, SW3 + Drainage2) where

Infiltratel is the infiltration at the surface

Drainagel is the drainage through soil layer 1

SW1 is the soil water content of soil layer 1

WO 2018/107245

PCT/AU2017/051403

Drainage2 is the drainage through soil layer 2 (which is below soil layer 1)

SW2 is the soil water content of soil layer 2

Drainage3 is the drainage through soil layer 3 (which is below soil layer 2)

SW3 is the soil water content of soil layer 1

FC1, FC2 and FC3 are field capacity of soil layers 1, 2 and 3 respectively [0112] In one example, potential evaporation may be determined as:

Potential evaporation_soii = Epan x (1 — Surface_C0ver)

Surf ace _cover 1 (1 Trans_covef) x (1 Dead_covef)

Dead_cover = min(l, (Litter_yi_d + Stdn_dead)/1000)

Pot_trans = EpanxTrans_cover

Trans_cover = 1 - exp(Green_sdmxlog(0.5)/gyld_covSO) where

Epan is pan evaporation

Surface_C0ver is fractional cover

Transcover is Term to convert fractional (green) into the transpiration rate modifier

Dead_cow,_r is Term to convert fractional (dead) into the transpiration rate modifier

Pot_trans is potential transpiration rate of a grass sward or crop [0113] In one example, the wail water supply may be determined as:

WO 2018/107245

PCT/AU2017/051403

SW1 - WP1

Sw supply _ratiol = max(0, _)

SW2 - WP2

Sw_supply ratio2 = max(0,)

SVK3 - WP3

Sw supply ratio3 = max(0,)

- ^y~ FC3-WP3^J

SWI1 = (1 + sin((Sw_supply_ratiol — 0.5) x pi)) x 0.5

Sl+72 = (1 + sin((Sw_supply_ratio2 — 0.5) x pi)) x 0.5

SWI3 = ^1 — cos (sW_supply_ratio3 ^xy)) ^χ roots_layer3

Total_swi = SW1 +SW2 + SW3

Profile_swi = min(l,Total_swi) where

WP1, WP2 and WP3 are the wilting points of plants accessing soil water at the respective layers (layers 1, 2 and 3).

rootsiayers is the proportion of roots in respective layers

Profile_swi is the soil water ratio for the whole profile [0114] In one example, potential transpiration may be determined as:

Trans = Pot_trans x Profile_swi

For layer 1, if Total_swi >0 then cSWl=Trans x SW1 /Total swi

WO 2018/107245

PCT/AU2017/051403 otherwise cSWl=0

For layer 2, if Total_swi >0 then cSW2=Trans x SW2/Total_swi otherwise cSW2=0

For layer 3, if Total_swi >0 then cSW3=Trans x SW3/Total_swi otherwise cSW3=0 [0115] In one example, soil evaporation may be determined as:

Esoil = Pot_esoiiX max(0, max(Es_swi_L12, Es_swi_L 1) ))

Es_swi_Ll = 0.285 x((SWl-ZDl)/(FCl-frDl)) (1-0.715 X (SW1-AD1/FC1 - ADI)

Es_swi_L12 = 0.117 x((SW1+SW2-AD1-WP2)/(FC1+FC2-AD1-WP2)) (1-0.0.833 X ((SW1+SW2-AD1-WP2)/(FC1+FC2 - AD1-WP2))

Es_swi_L2 = max(0,Es_swi_Ll2 - Es_swi_Ll)

For layer 1, if Esoil >0 then evaporation at layer 1 = Esoil x Es_swjM(m&x(Q,m2tx(Es_swiL12,Es_swjM)y) evaporation at layer 2 = Esoil x Es_swJE2l(m&x(Q,m&x(Es_swiL12,Es_swjM)y) where

ADI is a parameter to define the tension at which the water is held on the soils matrix

WO 2018/107245

PCT/AU2017/051403

Plant growth [0116] Plant growth may be determined by inspection of individual plants in the specified area 3. However, this may be time consuming and subjective thereby leading to inaccuracies and inconsistencies.

[0117] Therefore is some examples, the plant growth may be determined by sensor data that is indicative of plant growth that may include the environmental conditions of the specified area 3, during the preceding time period that may indicative of the plant growth. For example, favourable conditions for plants may be indicative of favourable plant growth. This model may be supplemented by and calibrated with other data objective data, such as data from the plant growth sensors described above.

[0118] The plant growth may be determined as gross primary production (GPP) or net primary production (NPP). This may be determined as the assimilation of carbon by plant growth. This may require determination of factors that affect plant growth including available soil water (that may include the results of determining soil water content described above), temperature and the associated constraints on photosynthesis, and use of radiation by the plant (which may include determination of light to plants in the specified area based exposure to sunlight, time of year, etc.). Agricultural practices such as grazing pastures and harvesting crops also modify plant growth through removal of the growth apparatus (leaves and stems) of the plant.

Example 1 of determining plant growth (Eco-physiological model driven by climate data and sensor data) [0119] One example of determining the plant growth includes using an eco-physiological model based on climate data. Climate data may include information such as rainfall, temperatures, wind, etc. that may be sourced from a meteorological station or climate data source 9. This may also take into account remotely sensed information about the status of radiation use by the plant (e.g. by reference to the leaf area index (LAI)).

[0120] In one example, the model may be based on the following equation for Gross Primary Production (GPP):

WO 2018/107245

PCT/AU2017/051403

GPP = α(ρ(1 - [D](T)(IB) Equation (1) [0121] The left set of terms describe energy use-carbon assimilation, where a is the maximum pasture or crop production governed by nutrient levels and plant physiology, Q is photosynthetically active solar radiation derived from monitored radiometer data (i.e. light sensor data), I is the canopy area term derived from remotely sensed NDVI and v is stomatai function.

[0122] The rate of energy use-carbon assimilation is limited firstly by the thermal environment described by the term T. T is a function of the daily temperature regime where monitored maximum and minimum temperatures are constrained with crop and pasture specific parameters defining the minimum, maximum and optimum response thresholds as well as the photosynthetic pathway (for C3 or C4 species). T also includes a temperature damage term which limits carbon assimilation under frost conditions for some species.

[0123] The rate of energy use-carbon assimilation is also modified by water limitation (W) which is a ramp function based on plant available water (PAW). This describes the familiar allometric response of crops and pastures to water availability. PAW is obtained in the water balance described below. PAW is obtained from a three layer force-restore water balance based on Rickert et al. (Rickert, K.G., Stuth, J.W. and McKeon, G.M. (2003). Modelling Pasture and Animal Production. In “Field and Laboratory Methods for Grassland and Animal Production Research” pp 29-66. CAB International, Wallingford).

[0124] The above described model may be further refined in an “assimilation mode” whereby parameters of the model may be adjusted by remotely sensed information. For remote sensed information about the actual status of radiation received at the specified area 3 may be used in place of, or to calibrate relevant parameters, of the leaf area index (LAI) used in the equation above. As remotely sensed information provides a census of landscape condition the determination of growth is also sensitive to the agricultural practices which modify the plant growth apparatus. This negates the need to survey cropping practices and stocking rates, which would be expensive and time consuming.

WO 2018/107245

PCT/AU2017/051403

Example 2 of determining plant growth - Crop development [0125] The GPP equation above may also contain an additional term for calculation of carbon assimilation of crops. This describes the development [£>] of annual wheat based on the work of Angus et al. (Angus J, Mackenzie D, Morton R and Schafer C (1981). Phasic development in field crops II. Thermal and photoperiodic responses of spring wheat. Field Crops Research 4, 269-283) , with the structure defined by Lui (Liu, D.L., 2007. Incorporating vernalization response functions into an additive phenological model for reanalysis of the flowering data of annual pasture legumes. Field Crops Research, 101: 331— 342) and the crop development function in APSIM. The development coefficient [£>] is a function of temperature and photoperiod through germination (G), emergence to floral initiation (SI), pre-flowering (S2) and flowering to harvest (S3) stages. G is determined by NDVI changes in the months April-June, while harvest is undertaken at 18 December thereby resetting the [D] value to 0 annually. The harvest date assumption is suitable when constructing drought indicators which are tracking potential climate limitations, but a specific date can be estimated to improve the prediction actual yields. It is to be appreciated that [D] may be selected for characteristics of other plants (or combination of other plants). In some examples, D may represent idealised models of a collection of plants.

[0126] In further examples the model, at least in part, may be based on crop harvest. For example, actual crop harvest may be recorded in the database 11, and this information may be recorded to verify and/or optimise the model parameters.

[0127] It is to be appreciated that the plant growth may be a combination of the features described in the above examples of determining plant growth.

Rainfall [0128] The determined rainfall is indicative of the rainfall at the specified area 3. In some examples, this includes rainfall over a specified time period since for practical purposes it is not the rainfall over a given hour or day that is relevant, but for a longer time period.

[0129] In some examples, the rainfall is provided as a percentile rank of daily rainfall aggregated over a twelve month summation period. From herein the summation period is

WO 2018/107245

PCT/AU2017/051403 termed the ‘aggregation window’. On a given day, the previous 12 months’ (365 days) data are aggregated then ranked within the baseline range (for example historical rainfall data obtained from 1985-2015). This is repeated for every day from 1915 to build an historical data base of 100 years. Thus the rainfall may be expressed as a value between 0 and 100, where, for any given climatic environment, values approaching 0 are close to the driest on the historical record and those approaching 100 are close to the wettest.

Developmental data, calibration parameter estimation of models [0130] In some examples, determining the soil water content, plant growth and/or rainfall involves modelling of the specified area 3.

[0131] To develop and calibrate the model, this may include using inputs such as using MODIS derived gross primary productivity data and/or MODIS derived actual evaporation data (which may in themselves be outputs of a model). This may also include field based data such as crop yield measurements, pasture yield growth measurements and soil water data from soil water sensors and sensor data from other remote sensors described above. Thus development of the model may include a technique of data assimilation. This developmental data may include historical data described above.

[0132] The model may also be calibrated for the specified area 3. This may include using historical data associated with a specified area 3, so that the model(s) are calibrated with parameters relevant local conditions and factors relevant to the specified area 2.

[0133] To improve the accuracy of the model, a technique of data assimilation is used and embedded into the model so that the model can be optimised with an automated procedure so that parameters can be continually tuned to improve accuracy. This may include tuning parameters by receiving additional information to calibrate and optimise the models. Such additional information may include sensor data, field data, climate data etc., as the system 1 is in use.

[0134] However, in some circumstances parameters may need to be estimated (for example if no sensor data is available for a specified area). In some examples, this may include

WO 2018/107245

PCT/AU2017/051403 providing parameters that are regionally based as a substitute for such parameters. An example of parameter estimation will now be described.

[0135] The regional algorithms have been developed to address a number of known challenges in the application of spatial agro-meteorological models. Firstly the spatial estimation of soil and plant functional parameters is problematic. Farm scale systems models require of a-priori estimates of plant functional form and soil physics parameters, either by measurement, field scale calibration procedures or expert interpretation. These procedures may be field and time intensive limiting the practical transportability of these models. While this is possible in localized case studies at single sites or regions, estimation across large regions has been problematic. In one example, New South Wales (Australia), the soil landscape variation is significant and there are broad ecological transitions defining the regional distribution of crop and pasture production. Pasture communities grade from subtropical in the north with greater prevalence of C4 grasses through to western land area where rangelands pastures dominate, through to temperate C3 grasslands. In the high rainfall zone micro-meteorological variation is significant given the effects of aspect and altitude.

[0136] An automated optimisation procedure 63 was developed to solve the modelling framework and yield a set of state wide parameters as illustrated in the example in Fig. 6. An objective function was formed as the weighted sum of squares (taking the Jacobian) between simulated values and the GPP values from 2000-2012, derived from the application of the diffuse model (see Donohue, Randall; Hume, Iain; Roderick, Michael; McVicar, Tim; Beringer, Jason; Hutley, Lindsay; et al. Evaluation of the remote-sensing-based DIFFUSE model for estimating photosynthesis of vegetation. Remote Sensing of Environment. 2014; 155:349-365). This was minimised initially by a pattern search algorithm to develop a trust region, then by a partial analytical method where the Jacobian was derived by finite differences. Initial guesses for the soil parameters were provided by the Australian Soil Landscape Grid (Viscarra Rossel et al. 2015, Soil and Landscape Grid of Australia, Soil Research, 53, pp 835-844) and published vegetation responses (see Nix, H.A. 1981. Simplified simulation models based on specified minimum data sets: the CROPEVAL concept. In: A. Berg (ed), Application of Remote Sensing to Agricultural Production Forecasting, Commission of the European Communities, Rotterdam, pp 151-169). The procedure was embedded as a large parallel processing experiment on a high performance

WO 2018/107245

PCT/AU2017/051403 computing mainframe, yielded a set of temperature, soil water and radiation use efficiency parameters across the region (New South Wales).

[0137] Independent verification of the model and parameters may be carried out by comparing optimised output from the models against field observations collected at 41 pasture production experiments across New South Wales. The summary results 65, where values have been aggregated to monthly mean growth values are provided in Fig. 7 where results close to line 67 indicate close correlation between the predictions from the model to the observed results.

Determining 120 the drought direction [0138] The drought direction indicates a trend in precipitation based on the determined rainfall. An example of determining 120 a drought direction will now be described. For each day of the determined rainfall a robust linear regression, which is not overtly sensitive to outliers, is fitted to the previous 90-days, and the slope of this function retained to form the drought direction. The value is rescaled to a range between -100 and +100 to visualise the drought direction. The key information is not the magnitude of the DDI across this scale, but the sign of the drought direction. If the value is negative there is a drying trend, and if it is positive the area is getting wetter.

[0139] In some examples, the drought direction is used as a categorical index only.

[0140] The drought direction provides information about trends in seasonal climate for the ‘Warning’ and ‘Recovery’ phases (e.g. early recovery and strong recovery). It was developed because early trial and error tests highlighted issues in reliably distinguishing between the ‘warning’ environmental condition from the two ‘recovery’ phases. In particular, for “warning” and “recovery, there may be circumstances where the rainfall, soil water content, and plant growth may be similar, but where the trend in environmental condition is trending towards different conditions. Accordingly, the drought direction allows evaluation of the trend and to distinguish between conditions that is in “warning” versus the “recovery” environmental conditions.

WO 2018/107245

PCT/AU2017/051403

Determining 130 indicators for drought conditions, normal condition, and recovery conditions [0141] Determining the various environmental conditions require the processing device to evaluate various inputs, in particular the soil water content, plant growth, rainfall and drought direction. The absolute values for these metrics may be normalised to assist in this evaluation process. This may also assist in improving accuracy of this determination normalising these values, for the specified area, would provide a more relevant indication of the state of the specified area 3.

[0142] As described above the soil water content (and rainfall) may be expressed as a value between 0 and 100 whereby the 0 is the driest and 100 is the wettest for that specified area 3. Similarly for plant growth, 0 may indicate the lowest plant growth and 100 may indicated the highest plant growth.

[0143] The method may then assist in identifying the indicators of respective conditions. This may be based on the percentile or percentile bands for various conditions. For example, an indicator of “normal conditions” for that metric may include values that are at or above a normal threshold (an in this example, the 50^th percentile). That is:

• Soil water content > 50:

• Plant growth >50;

• Rainfall > 50;

[0144] This normal condition may, of course, be expressed as a percentile band. For example where normal conditions are between the 50^th and 100^th percentile. It is to be appreciated that for each of the metrics, a different value or range for the “normal threshold” may be used. For example, the normal threshold for soil water content and rainfall may be the 50^th percentile, whilst the plant growth normal threshold may be the 60^th percentile.

[0145] In one example, an indicator of “drought conditions” for each metric may include values that are at or below a drought threshold (and in this example, the 10^th percentile). This may also be expressed as a percentile band between 0 and 10^th percentile for drought conditions.

WO 2018/107245

PCT/AU2017/051403 [0146] In another example, during recovery conditions the values may be between the drought conditions and normal conditions. Continuing from the above example values, this would be between the 10^th percentile and 50^th percentile. However it may be desirable to have further granularity within recovery conditions (i.e. to differentiate between early and strong recovery). Therefore an early recovery condition may be between the drought threshold (e.g. 10^th) and an intermediate recovery threshold (for example the 30^th percentile). A strong recovery condition may be between the intermediate recovery threshold and the normal threshold (e g. between the 30^th and 50^th percentile).

[0147] The above mentioned thresholds and percentile bands are examples. It is to be appreciated that the actual threshold value and/or bands may be different in other examples. Furthermore, it is to be appreciated that in some examples, “greater than” or “less than” the threshold may be adapted to “greater than or equal to” or “less than or equal to” and vice versa.

[0148] The above mentioned indicators of normal conditions, drought conditions, and intermediate recovery thresholds and/or percentile bands may be determined based on historical data. In some examples, these indicators may be specified (such as by an individual or organisation). Thus these indicators may be stored in the database 11 and determining the indicators may include receiving these indicators from the database 11. In yet other examples, these indicators may be determined by receiving the indicators, via the communications network 15, from another node in the network.

[0149] In the above examples, the determined indicators are based on a percentile of normalised data. However, it is to be appreciated that in other examples, the absolute values may be used for the metrics. For example the rainfall metric may be expressed in average millimetres per month over a preceding 12 month period. Accordingly, the indicator may be determined and expressed with a corresponding base. For example, the threshold for normal conditions may be where the average is 60mm per month or above, and the drought threshold may be where the average is 10 mm per month or below. It is to be appreciated that such threshold values may be determined based on historical data for each specified area 3 (or for a region including a plurality of specified areas). Such determination may be advantageous as these absolute values may be different for temperate regions versus, for example, semi-arid regions.

WO 2018/107245

PCT/AU2017/051403

Determining 140 the environmental conditions [0150] The above metrics may be evaluated by the processing device 13 to determine an environmental condition of the specified area 3.

[0151] Traditionally, “drought” conditions are periods when a rainfall deficit is driving a downturn in agriculture—relative to long term variability experienced in a given region—to a state where very little or no production is occurring, “non drought” conditions are those times when lack of rainfall is clearly not adversely impacting on production.

[0152] However, there may be technical difficulties determining the environmental condition between “drought” and “non drought”. Referring to Figs. 3 and 4, a “warning” environmental condition is where there is a detectable downward (drying) trend and conditions are starting to limit agricultural productivity. The two recovery phases occur when there is a detectable improving trend in environmental conditions. “Early recovery” signifies that a meteorological event has lifted conditions outside the range of what is considered a “drought”. “Strong recovery” signifies that conditions have continued to improve and production is building toward more optimal ‘non drought” conditions.

[0153] Examples of determining the various environmental conditions will now be described.

Warning 33 environmental condition [0154] The warning environmental condition is between drought and non drought conditions. Accordingly, the processing device 13 may determine this to occur where any one of the following metrics are between the drought threshold and normal threshold. That is:

• Drought threshold (e.g. 10^th percentile) < Soil water content < normal threshold (e.g. 50^th percentile): or • Drought threshold (e.g. 10^th percentile) < Plant growth < normal threshold (e.g. 50^thpercentile); or

WO 2018/107245

PCT/AU2017/051403 • Drought threshold (e.g. 10^th percentile) < Rainfall < normal threshold (e.g. 50^thpercentile);

[0155] However, the warning environmental condition is also indicative of a downward trend. Therefore, in addition, the processing device 13 must also determine that the drought direction is indicating a trend towards precipitation deficit. In one example (where a negative number indicates a drying trend), this may be expressed as:

• Drought direction < 0 [0156] The advantage of incorporating the drought direction is that the processing device 11 can differentiate the warning environmental condition from the recovery environmental conditions where such latter conditions may have metrics that overlap with the warning environmental condition. This is important as this result affects the response action of a farmer or other stakeholder.

Early recovery 37 environmental condition [0157] The early recovery 37 environmental condition is between drought and non drought environmental conditions that are starting to improve and out of drought (but may have a chance of re-entering drought). Accordingly, the processing device 13 may determine this to occur where all of the following metrics are between the drought threshold and intermediate recovery threshold. That is:

• Drought threshold (e.g. 10^th percentile) < Soil water content < intermediate recovery threshold (e.g. 30^th percentile): and • Drought threshold (e.g. 10^th percentile) < Plant growth < intermediate recovery threshold (e.g. 30^th percentile); and • Drought threshold (e.g. 10^th percentile) < Rainfall < intermediate recovery threshold (e.g. 30^th percentile);

[0158] However, the warning environmental condition is also indicative of an improving trend and where rainfall has occurred. Therefore, in addition, the processing device 13 must

WO 2018/107245

PCT/AU2017/051403 also determine that the drought direction is indicating a trend towards precipitation surplus.

In one example, this may be expressed as:

• 0 < Drought direction

Strong recovery 39 environmental condition [0159] The strong recovery 39 environmental condition is between drought and non drought environmental conditions but strongly indicating an improvement towards non drought conditions. Accordingly, the processing device 13 may determine this to occur where all of the following metrics are between the intermediate recovery threshold and normal threshold. That is:

• Intermediate recovery threshold (e.g. 30^th percentile)< Soil water content < normal threshold (e.g. 50^th percentile): and • Intermediate recovery threshold (e g. 30^th percentile)< Plant growth < normal threshold (e.g. 50^th percentile); and • Intermediate recovery threshold (e.g. 30^th percentile)< Rainfall < normal threshold (e.g. 50^th percentile):;

[0160] In addition, the processing device 13 must also determine that the drought direction is indicating a trend towards precipitation surplus. In one example, this may be expressed as:

• 0 < Drought direction

Drought 35 environmental condition [0161] The drought environmental condition is where any of the metrics indicate a drought condition. Accordingly, the processing device 13 may determine this to occur where any one of the following metrics are at or below the drought threshold. That is:

• Soil water content < Drought threshold (e.g. 10^th percentile): or

WO 2018/107245

PCT/AU2017/051403 • Plant growth < Drought threshold (e.g. 10^th percentile); or • Rainfall < Drought threshold (e.g. 10^th percentile);

[0162] Since the specified area 3 is in a drought condition, the processing device 11 may not need to take into account the drought direction.

Non-dr ought 31 environmental condition [0163] The non drought 31 environmental condition is where all of the metrics indicate that the specified area is in a non drought condition. Accordingly, the processing device 13 may determine this to occur where all one of the following metrics are at or above the normal threshold. That is:

• Normal threshold (e.g. 50^th percentile) < Soil water content: and • Normal threshold (e.g. 50^th percentile) < Plant growth; and • Normal threshold (e.g. 50^th percentile) < Rainfall;

[0164] Since the specified area 3 is in a non drought condition, the processing device 11 may not need to take into account the drought direction.

Sending 150 a notification based on the determined environmental condition [0165] The processing device 13, in response to determining one or more of the environmental conditions sends a notification over a communications network to a communication device to indicate the environmental condition. For example after determining a warning environmental condition, the processing device may send a notification to a stakeholder of a specified area 3 (such as a farmer) to alert them that the specified area in a warning environmental condition. This may allow the stakeholder to take action to ameliorate the effects of the environmental condition (for example diverting water from alternative sources).

WO 2018/107245

PCT/AU2017/051403 [0166] In other examples, the notification may be sent based on a change of environmental condition. For example a change from a drought condition to early recovery condition may trigger a notification to a stakeholder that conditions are improving. For example, this change may alert a farmer to being sowing crop.

[0167] In other examples, the notifications may be sent to a communications device that is in communication with machinery and equipment. For example, the notification may be used to activate or deactivate irrigation equipment based on the determined environmental condition. In another example, the notification may be used to activate or deactivate livestock feeding equipment.

[0168] In some examples, the notification may be generated on a display associated with the communication device 17. In some examples, the notification may be generated graphically on a map overlay. Fig. 8 shows a map of a region (New South Wales) with notifications of environmental conditions that are overlayed. For example, specified areas that are in warning, early recovery, strong recovery environmental conditions have respective differentiating colours. Areas in drought and non drought may also have respective colours. It is to be appreciated patterns or other visual differentiators may be used to indicate respective notifications.

Spatial aggregation [0169] In some examples, privacy or confidential information concerns may require adjustment in the granularity of the data, yet still require accurate representation of the data for a particular area. For example, in some examples the specified area may be calculated at a 5 km by 5 km grid, and this resolution may allow third parties (other than say, a farmer of that area) from obtaining commercially sensitive information on the environmental condition of that land (for example, using the information to derive information on specific productivity of that land that may be affected by other factors). Accordingly, in some examples where the environmental conditions may be sent publicly to third parties, the information in relation to the environmental condition of the specified area 3 may be aggregated with other nearby areas 3.

WO 2018/107245

PCT/AU2017/051403 [0170] For example, some governments classify areas by “parish”. Such a parish may contain multiple specified areas 3 in the vicinity of each other. Thus in some examples, sending notification of the environmental condition 2 may include sending information in relation to the geographic parish that the specified area 3 is located within This may be useful for public reporting to maintain privacy and confidentiality.

[0171] An example of spatial aggregation will now be described with reference to Fig. 9. This shows a subset 53 of a geographic region 51 whereby the subset is divided into grids 55 for modelling and calculation purposes. A parish 57 partially overlaps a plurality of the grids 55. In this example the grids 55 each represent a respective specified area 3 in which the environmental conditions and metrics are individually calculated).

[0172] In this example, the environmental condition and metrics (of soil water content, plant growth, and rainfall) are aggregated from the individually calculation scale (of 5km by 5 km) grids 55 to provide a result at a parish 57 level. These scales of aggregation are shown in Fig. 9, outlining the parishes in relation to the underlying grids 55. New South Wales (NSW) parishes provide a convenient reporting unit for a spatial process like drought, as they are fine enough to capture the spatial variability associated with major drought events across the State, but make it difficult to identify individual farms. This is an important requirement under privacy considerations where individual farm level information is not publicly disclosed. There are 7378 parishes in NSW with the larger parishes located to the west of the State reflecting sparser settlement patterns. Generally NSW parishes contain 4-5 grid cells (of 5km by 5km each), which for most of the State would contain 10-20 individual property holdings. The largest parish in the west of the State has 27 grid cells, while there are 300 parishes in the eastern Sydney metropolitan and urbanised coastal (non-agricultural) regions that only have 1 grid cell.

[0173] The algorithm performs aggregation by determining the grids 55 (i.e. a plurality of specified areas 3) that are associated with the parish 57. In some examples this association is determined by grids 55 that are overlapped wholly or in part by the parish 57.

[0174] In further examples, the aggregation algorithm finds the relevant centroids 59 of the underlying grid 55, with a 4km² buffer around the parish boundary to account for uncertainty

WO 2018/107245

PCT/AU2017/051403 in the underlying gridded data. That is, associating grids 55 that are proximal to the boundary of the geographic parish 57. This is shown in the example for the parish “Athol” 57 in Fig. 9.

[0175] To determine the environmental condition of the geographic parish 57, the processing device may determine the environmental condition of each of the associated grids 55 (specified areas 3) and determine the environmental condition that occurs the most frequently (i.e. the mode). This may then sent, over the communications network, as a notification of the environmental condition of the geographic parish 57. This aggregation at the parish level may alleviate privacy concerns.

[0176] In some examples, it may be desirable to calculate the metrics (such as the soil water content, plant growth, and rainfall) of the parish 57 to provide further information on the parish 57. In some examples, this calculation for the parish 57 may be performed by determining the average values for the metrics from the associated grids 55.

[0177] It is to be appreciated that modelling may be performed and repeated in other scales. For example the grids may be smaller (e.g. 2 metre by 2 metre grids) and then aggregated to a field (or other defined area) to support farm level decisions.

User interface and reporting [0178] Fig. 10 illustrates a display 1112 with a user interface 200 to provide a visual representation of the notification to a user. This includes a map 201 (in this particular example, the state of New South Wales, Australia), with overlay of different shades (or colours, or other markings) to indicate the environmental conditions of respective areas in the state. This includes a key 205 to show the shades/colours and the respective environmental condition. The darker shade in the middle to western area 207 on the map 201 indicates that these areas are in the drought environmental condition. A pie chart 209 shows the various percentages of the environmental condition for the state.

[0179] The map 201 of the state is further divided into smaller geographic regions. For example the “WESTERN”, “CENTRAL WEST”, “NORTH WEST” regions. The user interface 200 may allow a user to select one of these regions for more in depth (and granular)

WO 2018/107245

PCT/AU2017/051403 details. For example, the user may use a cursor 213 to select the CENTRAL WEST region 211.

[0180] Fig. 11 illustrates the display 1112 after selection of the CENTRAL WEST region 211. This illustrates the various environmental conditions within this region 211 with greater detail. A pie chart 219 is provided to show the percentages of the environmental conditions in the region 211.

[0181] Fig. 11 also illustrates the boundaries of various smaller sub regions, which in this case are shires. In Fig. 11, there are eleven shires in the CENTRAL WEST region 211. A user may select one of the shires for further detail, for example the PARKES shire 215.

[0182] Fig. 12 illustrates the display 1112 after selection of the PARKES shire 215. This shows the various parishes within the PARKES shire 215. In this example, each parish has a respective environmental condition (as represented by the shade/colour overlayed in the parish boundary). As noted above, the aggregation of environmental conditions to a parish level may be useful to alleviate privacy concerns where such information is publicly available.

[0183] Fig. 12 also illustrates a selection of the MINGELO parish 221 with the cursor 213. In response, further detail of the parish 221 is provided in a box 223. This includes the name of the parish and relevant information including the environmental condition (in this case a “warning” environmental condition), the drought direction, plant growth, rainfall and soil water content.

[0184] In some examples, information on the environmental conditions are hosted on a publicly accessible server. This may include a web portal to access this information and to generate the above mentioned user interface 200 at a respective communication device of a user.

[0185] In other examples, the information may be provided on a subscription basis, where a stakeholder makes contributions (directly, or indirectly) to receive the information on the environmental conditions. In some examples, the contribution may be financial and in other examples, the contribution may include information (such as data from remote sensors on a

WO 2018/107245

PCT/AU2017/051403 farm) for use by the system 1. In some examples, the information provided may be limited, for example, a farmer may subscribe to information related to their region or parish.

[0186] The notifications may be pushed to a communication device of the stakeholder. For example, a stakeholder may specify that they wish to be alerted to certain environmental conditions or changes to environmental conditions for specified areas. This may include a message with images of a map and information as discussed above. In some examples, this may include a message with a URL (uniform resource locator) to allow the user to access the above information from a server.

[0187] In some examples, the notification may be included in a report for the specified area 3, parish 57 or other geographic region. This may include the system generating a report with one or more of the above described maps and overlay. The report may also include the pie chart 209, 219 as well as details in relation to rainfall, soil water content and plant growth data. This report may be selectively sent to stakeholders in the respective specified areas 3.

[0188] In yet other examples, the notification and report may be produced at the farm, paddock or sub-paddock level. This may include receiving information from multiple sources, remote sensors 4 associated with the local region. This can then be supplemented by a farmer’s private data from their own remote sensors 4. Thus the farmer may be able to receive reports and notifications that are more specific to their use. This may include consideration of parameters such as type of usage, crops, livestock, and fertilizer (amongst other variables) for particular specified areas of their farm.

Variations [0189] It is to be appreciated that the soil water data, plant growth data and rainfall data may from the remote sensors 4 described above may be enhanced by other information in the process of data assimilation. Fig. 14 illustrates another example of the system 301 with other information sources. For example, climate models 303 may be simulated in a computer (such as a forecaster 305) to adjust, and/or predict future, soil water content, plant growth and rainfall.

WO 2018/107245

PCT/AU2017/051403 [0190] As an illustrative example, there may be a correlation between recent rainfall and increase in soil moisture content. However the remote sensors 4 and/or climate data source 9 collecting respective data may be sampling at different rates and different times. Thus there may be a temporal mismatch (i.e. lead/lag) between the data. Thus modelling may be used to assist in predicting or interpolating data.

[0191] In one simplistic example, this may include (based on historical data) determining a model that can estimate an increase in soil water content for a given amount of rainfall that is received in the specified area 3. This estimated increase may then be used in conjunction with the last soil water data from the soil water sensor 5 to determine up to date soil water content.

[0192] In further examples, the soil water data, plant growth data and rainfall data may be enhanced with information from field monitoring 307. For example, a farmer may observe characteristics indicative of plant growth and provide an input through a field monitoring device (such as a wireless communication device), and this information may be used to enhance the determined plant growth. For example, a farmer may observe crop yield and enter such information through the field monitoring device. This crop yield may be compared with historic yields and used as a factor to assist in determining plant growth.

[0193] In some examples, the processing device 13 may provide localised information to a private monitoring community 309 so that a farmer may receive more detailed information at the farm level (instead of, for example, the parish level)). This farm level reporting may provide information, for example, at a paddock level (even smaller).

Processing device [0194] The system 1 may include a computer that includes a processing device 1013 as shown in Fig 14. The communication device 17 may also include a processing device. The processing device 1013 includes a processor 1102 connected to a program memory 1104, a data memory 1106, a communication port 1108 and a user port 1110. The program memory 1104 is a non-transitory computer readable medium, such as a hard drive, a solid state disk or CD-ROM. Software, that is, executable program instructions that are stored on program memory 1104 causes the processor 1102 to perform the method 100 illustrated in Fig. 2.

WO 2018/107245

PCT/AU2017/051403 [0195] It is to be appreciated that in some examples, the processing device may include multi-core processors. In further examples, multiple processing devices may be configured for multiprocessing to perform the method.

[0196] The processor 1102 may receive data, such as sensor data or historical data from data memory 106 as well as from the communications port 108 and the user port 1110. The user port 1110 may also be connected to a display 1112 that shows a visual representation 1114 of the notification, which may include a map overlay, as well as forming a user interface to a user.

[0197] Although communications port 1108 and user port 1110 are shown as distinct entities, it is to be understood that any kind of data port may be used to receive data, such as a network connection, a memory interface, a pin of the chip package of processor 1102, or logical ports, such as IP sockets or parameters of functions stored on program memory 1104 and executed by processor 1102. These parameters may be stored on data memory 1106 and may be handled by-value or by-reference, that is, as a pointer, in the source code.

[0198] The processor 1102 may receive data through all these interfaces, which includes memory access of volatile memory, such as cache or RAM, or non-volatile memory, such as an optical disk drive, hard disk drive, storage server or cloud storage. The processing device 1013 may further be implemented within a cloud computing environment, such as a managed group of interconnected servers hosting a dynamic number of virtual machines.

[0199] It should be understood that the techniques of the present disclosure might be implemented using a variety of technologies. For example, the methods described herein may be implemented by a series of computer executable instructions residing on a suitable computer readable medium. Suitable computer readable media may include volatile (e g. RAM) and/or non-volatile (e.g. ROM, disk) memory, carrier waves and transmission media. Exemplary carrier waves may take the form of electrical, electromagnetic or optical signals conveying digital data steams along a local network or a publically accessible network such as the internet.

[0200] It should also be understood that, unless specifically stated otherwise as apparent from the following discussion, it is appreciated that throughout the description, discussions

WO 2018/107245

PCT/AU2017/051403 utilizing terms such as estimating or processing or computing or calculating, optimizing or determining or displaying or “maximising” or the like, refer to the action and processes of a computer system, or similar electronic computing device, that processes and transforms data represented as physical (electronic) quantities within the computer system's registers and memories into other data similarly represented as physical quantities within the computer system memories or registers or other such information storage, transmission or display devices.

[0201] It will be appreciated by persons skilled in the art that numerous variations and/or modifications may be made to the above-described embodiments, without departing from the broad general scope of the present disclosure. The present embodiments are, therefore, to be considered in all respects as illustrative and not restrictive.

Claims

CLAIMS:

1. An environmental monitoring system (1) for detecting an environmental condition of a specified area (3), the system comprising:

- one or more remote sensors (4) and/or a climate data source (9) to provide sensor data indicative of soil water content, plant growth and rainfall associated with the specified area (3);

- a database (11) to store historical data based on historical soil water data, historical plant growth data, and historical rainfall data; and

- a processing device (13) to:

- determine (110) soil water content, plant growth and rainfall based on sensor data from the one or more remote sensors (4) and/or climate data source (9) and historical data;

- determine (120) a drought direction associated with the specified area (3) that indicates a trend in precipitation based on the determined rainfall and historical rainfall data;

- determine (130) respective indicators for soil water content, plant growth and rainfall during drought conditions and normal conditions associated with the specified area (3) that are based on historical data;

- determine (140) a warning (33) environmental condition associated with the specified area (3) based on determination of:

- a drought direction indicating a trend towards precipitation deficit; and at least one of:

- determined soil water content is greater than the indicator for soil water content during drought conditions but less than the indicator for soil water content during normal conditions;

WO 2018/107245

PCT/AU2017/051403

- determined plant growth is greater than the indicator for plant growth during drought conditions but less than the indicator for plant growth during normal conditions; and

- determined rainfall is greater than the indicator for rainfall during drought conditions but less than the indicator for rainfall during normal conditions;

- based on determining a warning (33) environmental condition, send (150), over a communications network (15), a notification to a communication device (17) indicating a warning environmental condition associated with the specified area.
2. A system according to claim 1, wherein the processing device is further configured to:

- determine the indicators for soil water content, plant growth and rainfall during early recovery conditions associated with the specified area based on historical data;

- determine an early recovery (37) environmental condition associated with the specified area based on determination of:

- a drought direction indicating a trend towards precipitation surplus; and

- determined soil water content is within the indicator for soil water content during early recovery conditions; and

- determined plant growth is within the indicator for plant growth early recovery conditions; and

- determined rainfall is within the indicator for rainfall during early recovery conditions; and

- based on determining an early recovery environmental condition, send, over a communications network, a notification indicating an early recovery environmental condition associated with the specified area.

WO 2018/107245

PCT/AU2017/051403
3. A system according to claim 2, wherein the processing device is further configured to:

- determine the indicators for soil water content, plant growth and rainfall during strong recovery conditions (39) associated with the specified area based on historical data;

- determine a strong recovery (39) environmental condition associated with the specified area based on determination of:

- a drought direction indicating a trend towards precipitation surplus; and

- determined soil water content is within the indicator for soil water content strong recovery conditions; and

- determined plant growth is within the indicator for plant growth strong recovery conditions; and

- determined rainfall is within the indicator for rainfall during strong recovery conditions; and

- based on determining a strong recovery environmental condition, send, over a communications network, a notification indicating a strong recovery environmental condition associated with the specified area.
4. A system according to any one of the preceding claims, wherein the processing device is further configured to:

- determine a drought (35) environmental condition associated with the specified area based on determination of at least one of:

- determined soil water content is less than the indicator for soil water content during drought conditions;

- determined plant growth is less than the indicator for plant growth during drought conditions; and

WO 2018/107245

PCT/AU2017/051403

- determined rainfall is less than the indicator for rainfall drought conditions; and

- based on determining a drought environmental condition, send, over a communications network, a notification indicating a drought environmental condition associated with the specified area.
5. A system according to any one of the preceding claims, wherein the processing device is further configured to:

- determine a non drought (31) environmental condition associated with the specified area based on determination of:

- determined soil water content is within the indicator for soil water content during normal conditions; and

- determined plant growth is within the indicator for plant growth during normal conditions; and

- determined rainfall is within the indicator for rainfall during normal conditions; and

- based on determining a non drought environmental condition, send, over a communications network, a notification indicating a non drought environmental condition associated with the specified area.
6. A system according to any one of claims 1 to 5, wherein the processing device is further configured to determine the soil water content, plant growth and rainfall as an aggregation of respective sensor data from a specified preceding period.
7. A system according to any one of claims 1 to 6, wherein the processing device is further configured to:

- determine a range or threshold value as the indicator for soil water content in normal conditions;

WO 2018/107245

PCT/AU2017/051403

- determine a range or threshold value as the indicator for plant growth in normal conditions;

- determine a range or threshold value as the indicator for rainfall in normal conditions;

- determine a range or threshold value as the indicator for soil water content in drought conditions;

- determine a range or threshold value as the indicator for plant growth in drought conditions; and

- determine a range or threshold value as the indicator for rainfall in drought conditions.
8. A system according to any one of claims 2 to 7, wherein the processing device is further configured to:

- determine a range or threshold value as the indicator for soil water content in early recovery conditions;

- determine a range or threshold value as the indicator for plant growth in early recovery conditions;

- determine a range or threshold value as the indicator for rainfall in early recovery conditions;

- determine a range or threshold value as the indicator for soil water content in strong recovery conditions;

- determine a range or threshold value as the indicator for plant growth in strong recovery conditions; and

- determine a range or threshold value as the indicator for rainfall in strong recovery conditions.

WO 2018/107245

PCT/AU2017/051403
9. A system according to either claim 7 or 8 wherein the range or threshold value is expressed as a percentile range or percentile threshold value.
10. A system according to any one of claims 1 to 8, wherein the processing device is further configured to determine the soil water content, plant growth and rainfall as normalised data values based on historical soil water content, historical plant growth and historical rainfall.
11. A system according to any one of the preceding claims further configured to detect an environmental condition of a geographic parish (57) that includes an associated plurality of specified areas (3), wherein the processing device (13) is further configured to:

- determine an environmental condition for each of the associated plurality of specified areas (3);

- determine the environmental condition associated with the parish (57) based on the environmental condition that occurs most frequently in the associated plurality of specified areas (3), wherein to send, over the communications network (15), the notification to the communication device (17) includes the processing device configured to send the environmental condition associated with the geographic parish (57).
12. A system according to claim 11 wherein the associated plurality of specified areas (3) include specified areas that the geographic parish (57) overlap.
13. A system according to claim 12 wherein the associated plurality of specified areas (3) further include specified areas that are proximal to the boundary of the geographic parish (57).
14. A system according to any one of the preceding claims wherein the one or more remote sensors (4) comprise one or more of the following:

- a soil water sensor (5) to provide soil water data associated with the specified area (3);

WO 2018/107245

PCT/AU2017/051403

- a rain gauge (7) to provide rainfall data associated with the specified area (3);

- a plant growth monitoring device (7) to provide plant growth data associated with the specified area (3);

- a temperature sensor;

- a light sensor;

- an evaporation gauge;

- a humidity sensor;

- an anemometer; and

- a barometer.
15. A system according to any one of the preceding claims further comprising an aerial drone wherein the aerial drone has one or more of the remote sensors on board.
16. A system according to any one of the preceding claims wherein the climate data source (9) is associated with one or more of the remote sensors (4) that collect sensor data associated with the specified area.
17. A system according to any one of the preceding claims wherein the processing device is further configured to generate, at a display, a graphical representation of the notifications as an overlay of a map that includes at least the specified area (3).
18. A computer-implemented method for detecting an environmental condition of a specified area (3), the method comprising:

- determining (110) soil water content, plant growth and rainfall based on sensor data from one or more remote sensors (4) and/or climate data source (9) and historical data,

WO 2018/107245

PCT/AU2017/051403 wherein the one or more remote sensors and/or climate data source provides sensor data indicative of soil water content, plant growth and rainfall, and wherein the historical data is based on at least historical soil water data, historical plant growth and historical rainfall data;

- determining (120) a drought direction associated with the specified area (3) that indicates a trend in precipitation based on the determined rainfall and historical rainfall data;

- determining (130) respective indicators for soil water content, plant growth and rainfall during drought conditions and normal conditions associated with the specified area (3) that are based on historical data;

- determining (140) a warning (33) environmental condition associated with the specified area (3) based on determination of:

- a drought direction indicating a trend towards precipitation deficit; and at least one of:

- determined soil water content is greater than the indicator for soil water content during drought conditions but less than the indicator for soil water content during normal conditions;

- determined plant growth is greater than the indicator for plant growth during drought conditions but less than the indicator for plant growth during normal conditions; and

- determined rainfall is greater than the indicator for rainfall during drought conditions but less than the indicator for rainfall during normal conditions;

- based on determining a warning (33) environmental condition, sending (150), over a communications network (15), a notification to a communication device (17) indicating a warning environmental condition associated with the specified area.

WO 2018/107245

PCT/AU2017/051403
19. A method according to claim 18, the method further comprising:

- determining the indicators for soil water content, plant growth and rainfall during early recovery conditions associated with the specified area based on historical data;

- determining an early recovery (37) environmental condition associated with the specified area based on determination of:

- a drought direction indicating a trend towards precipitation surplus; and

- determined soil water content is within the indicator for soil water content during early recovery conditions; and

- determined plant growth is within the indicator for plant growth early recovery conditions; and

- determined rainfall is within the indicator for rainfall during early recovery conditions; and

- based on determining an early recovery environmental condition, sending, over a communications network, a notification indicating an early recovery environmental condition associated with the specified area.
20. A method according to claim 19, the method further comprising:

- determining the indicators for soil water content, plant growth and rainfall during strong recovery conditions (39) associated with the specified area based on historical data;

- determining a strong recovery (39) environmental condition associated with the specified area based on determination of:

- a drought direction indicating a trend towards precipitation surplus; and

- determined soil water content is within the indicator for soil water content strong recovery conditions; and

WO 2018/107245

PCT/AU2017/051403

- determined plant growth is within the indicator for plant growth strong recovery conditions; and

- determined rainfall is within the indicator for rainfall during strong recovery conditions; and

- based on determining a strong recovery environmental condition, sending, over a communications network, a notification indicating a strong recovery environmental condition associated with the specified area.
21. A method according to any one of claims 18 to 20, the method further comprising:

- determining a drought (35) environmental condition associated with the specified area based on determination of at least one of:

- determined soil water content is less than the indicator for soil water content during drought conditions;

- determined plant growth is less than the indicator for plant growth during drought conditions; and

- determined rainfall is less than the indicator for rainfall drought conditions; and

- based on determining a drought environmental condition, sending, over a communications network, a notification indicating a drought environmental condition associated with the specified area.
22. A method according to any one of claims 18 to 21, the method further comprising:

- determining a non drought (31) environmental condition associated with the specified area based on determination of:

- determined soil water content is within the indicator for soil water content during normal conditions; and

WO 2018/107245

PCT/AU2017/051403

- determined plant growth is within the indicator for plant growth during normal conditions; and

- determined rainfall is within the indicator for rainfall during normal conditions; and

- based on determining a non drought environmental condition, sending, over a communications network, a notification indicating a non drought environmental condition associated with the specified area.
23. A method according to any one of claims 18 to22 further comprising generating, at a display associated with the communications device, a graphical representation of the notification as an overlay on a map that includes at least the specified area.
24. Software that, when installed on a computer, causes the computer to perform the method according to any one of claims 18 to 23.